Metabolomic Alterations in Methotrexate Treatment of Moderate-to-Severe Psoriasis

Abstract

Background
Aberrant lipid metabolism alterations in skin tissue, blood, or urine have been implicated in psoriasis. Here, we examined lipid metabolites related to psoriasis and their association with the age of disease onset.

Material/Methods
Differences in lipid metabolites before and after methotrexate (MTX) treatment were evaluated. The discovery cohort and validation cohort consisted of 50 and 46 patients, respectively, with moderate-to-severe psoriasis. After MTX treatment, the patients were divided into response (Psoriasis Area and Severity Index [PASI] 75 and above) and non-response (PASI below 75) groups, blood was collected for serum metabolomics, and multivariate statistical analysis was performed.

Results
We detected 1546 lipid metabolites. The proportion of the top 3 metabolites was as follows: triglycerides (TG, 34.8%), phospholipids (PE, 14.5%), phosphatidylcholine (PC, 12.4%); diglycerides (DG) (16: 1/18: 1), and DG (18: 1/18: 1) showed strong positive correlations with onset age. There were marked changes in TG (16: 0/18: 0/20: 0), TG (18: 0/18: 0/22: 0), TG (14: 0/18: 0/22: 0), TG (14: 0/20: 0/20: 0), lysophosphatidylcholine (LPC) (16: 0/0: 0), LPC (18: 0/0: 0), LPC (14: 0/0: 0), and LPC (18: 1/0: 0) levels before and after 12 weeks of MTX treatment. The glycerophospholipid metabolic pathway was implicated in psoriasis development. Of the 96 recruited patients, 35% were MTX responders and 65% non-responders. PE (34: 4) and PE (38: 1) levels were significantly different between the groups. Obvious differences in lipid metabolism were found between early-onset (<40 years) and late-onset (≥40 years) psoriasis. Significant changes in serum lipid profile before and after MTX treatment were observed.

Conclusions
The specific lipid level changes in responders may serve as an index for MTX treatment efficacy evaluation.

Full text

Received: 2023.12.01
Accepted: 2024.02.22
Available online: 2024.03.12
Published: 2024.05.08
3584 1 4 30
Metabolomic Alterations in Methotrexate
Treatment of Moderate-to-Severe Psoriasis
ACDE 1-3 Yi Zhou
BCDE 1-3 Yantao Ding
F 4 Mengxing Cui
BCD 1-3 Yuanjing Zhang
B 2,3 Mengwei Wang
F 5 Feiran Zhou
F 6 Yi Su
ACDG 1-3 Bo Liang
ACD 1-3 Fusheng Zhou
Corresponding Author: Fusheng Zhou, e-mail: biozhoufs@163.com, Bo Liang, e-mail: anyiliangbo@vip.126.com
Financial support: This study was funded by the Natural Science Research Project of Higher Education Department of Anhui Province (No. KJ2021A0284);
Fund of Anhui Provincial Institute of Translational Medicine (No. 2021zhyx-C31); Natural Science Foundation Project of Anhui Province
(No. 2308085MH263); and Innovation and Entrepreneurship Training Program of Anhui Medical University (No. S202210366027
and No. S20221366060)
Conflict of interest: None declared
Background: Aberrant lipid metabolism alterations in skin tissue, blood, or urine have been implicated in psoriasis. Here, we
examined lipid metabolites related to psoriasis and their association with the age of disease onset.
Material/Methods: Differences in lipid metabolites before and after methotrexate (MTX) treatment were evaluated. The discovery
cohort and validation cohort consisted of 50 and 46 patients, respectively, with moderate-to-severe psoriasis.
After MTX treatment, the patients were divided into response (Psoriasis Area and Severity Index [PASI] 75 and
above) and non-response (PASI below 75) groups, blood was collected for serum metabolomics, and multivar-
iate statistical analysis was performed.
Results: We detected 1546 lipid metabolites. The proportion of the top 3 metabolites was as follows: triglycerides
(TG, 34.8%), phospholipids (PE, 14.5%), phosphatidylcholine (PC, 12.4%); diglycerides (DG) (16: 1/18: 1),
and DG (18: 1/18: 1) showed strong positive correlations with onset age. There were marked changes in TG
(16: 0/18: 0/20: 0), TG (18: 0/18: 0/22: 0), TG (14: 0/18: 0/22: 0), TG (14: 0/20: 0/20: 0), lysophosphatidylcho-
line (LPC) (16: 0/0: 0), LPC (18: 0/0: 0), LPC (14: 0/0: 0), and LPC (18: 1/0: 0) levels before and after 12 weeks
of MTX treatment. The glycerophospholipid metabolic pathway was implicated in psoriasis development. Of
the 96 recruited patients, 35% were MTX responders and 65% non-responders. PE (34: 4) and PE (38: 1) levels
were significantly different between the groups. Obvious differences in lipid metabolism were found between
early-onset (<40 years) and late-onset (³40 years) psoriasis. Significant changes in serum lipid profile before
and after MTX treatment were observed.
Conclusions: The specific lipid level changes in responders may serve as an index for MTX treatment efficacy evaluation.
Keywords: Lipidomics • Methotrexate • Psoriasis
Full-text PDF: https://www.medscimonit.com/abstract/index/idArt/943360
Authors’ Contribution:
Study Design A
Data Collection B
Statistical Analysis C
Data Interpretation D
Manuscript Preparation E
Literature Search F
Funds Collection G
1 Department of Dermatology and Venereology, The First Affiliated Hospital,
Anhui Medical University, Hefei, Anhui, PR China
2 Institute of Dermatology, Anhui Medical University, Hefei, Anhui, PR China
3 Key Laboratory of Dermatology, Anhui Medical University, Ministry of Education,
Hefei, Anhui, PR China
4 Department of Clinical Laboratory, The First Affiliated Hospital, Anhui Medical
University, Hefei, Anhui, PR China
5 The First Clinical Medical School, Anhui Medical University, Hefei, Anhui, PR China
6 The Second Clinical Medical School, Anhui Medical University, Hefei, Anhui,
PR China
e-ISSN 1643-3750
© Med Sci Monit, 2024; 30: e943360
DOI: 10.12659/MSM.943360
Publisher’s note: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated
organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be
made by its manufacturer, is not guaranteed or endorsed by the publisher
e943360-1
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

Introduction
Psoriasis is a common chronic inflammatory skin disease char-
acterized by disruption of epidermal homeostasis and impair
-
ment of skin barrier function [1]. Psoriasis is a common health
issue worldwide, and approximately 125 million people are af-
fected. The global prevalence of psoriasis varies by geograph-
ical location. In general, psoriasis is more common in colder
regions than in tropical regions. The prevalence of psoriasis in
China is approximately 0.47%, and psoriasis affects individu-
als of both sexes almost equally [2].
Psoriasis can occur at any age and has a bimodal age at onset
(16-22 and 55-60 years). According to the age of onset, psori-
asis can be divided into 2 clinical subtypes, early-onset (EOP
<40 years) and late-onset (LOP ³40 years) psoriasis [3]. EOP
is generally associated with HLACw6, -B13, and -B57, where-
as LOP is correlated with HLA-Cw2 and -B27 [4]. EOP is more
serious and likely to relapse than LOP. However, LOP poses a
heavier and more rapid inflammatory burden. Cardiovascular
diseases, diabetes mellitus, hypertension, and metabolic syn-
dromes are more frequent in LOP than in EOP [5].
Psoriasis is a systemic and life-long disease and is often accom-
panied by other conditions, such as hyperlipidemia, hyperten-
sion, type 2 diabetes, obesity, and cardiovascular complications,
which can be categorized as metabolic syndrome. The more se-
vere the psoriasis condition, the higher the risk of the metabolic
syndrome. However, the link between them is not clear [6]. In re-
cent years, accumulating evidence has shown that lipid expres-
sion and metabolic disorders are more commonly observed in
the skin tissue, serum, and urine of patients with psoriasis [7].
The treatment of psoriasis can be roughly divided into 2 cate-
gories: topical and systemic. Topical medication mainly refers
to local treatment or phototherapy for mild psoriasis, while
systemic treatment mainly involves the use of immunosup-
pressants, such as methotrexate (MTX) and cyclosporine. MTX
has been used as a first-line treatment for moderate-to-severe
psoriasis for more than 50 years. In recent years, although the
discovery and use of biologics have been widely accepted in
psoriasis treatment, MTX remains the most commonly used
systemic drug for psoriasis globally. MTX acts mainly through
the following mechanisms. MTX can (1) reduce DNA synthe-
sis and induce apoptosis of keratin-forming cells; (2) exert im-
munosuppressive effects through chemotaxis and reduction of
adenosine levels, suppressing the production of T and B lym-
phocytes, pro-inflammatory factors, neutrophils, and mono-
cytes; and (3) exert an anti-inflammatory effect by inhibiting
the JAK/STAT pathway at low doses [8]. However, the effec-
tiveness of MTX in the treatment of different types of psoriasis
varies. Compared with rich clinical experience, the quality of
clinical research on this drug is relatively poor. In our previous
study, MTX was more effective and showed fewer adverse ef-
fects in patients with psoriasis without arthritis than in those
with arthritis [9]. We examined the association between pso-
riasis susceptibility gene variants and clinical response to MTX
and found that patients with psoriasis with the TT genotype
rs10036748 in TNIP1 responded better to MTX [10]. These re-
sults suggest that the effectiveness of MTX is influenced by
multiple factors, such as medical conditions and genetic factors.
Lipidomics refers to the large-scale analysis and characterization
of a range of lipid species in biological systems. Considerable
changes in lipids, such as lysophosphatidic acid, lysophospha-
tidylcholine (LPC), phosphatidylinositol, phosphatidylcholine
(PC), and phosphatidic acid (PA) are observed in the plasma
of patients with psoriasis [11]. This observation suggests that
abnormal lipid metabolism may prevail in psoriasis, which has
important implications for disease pathogenesis, outcome, and
management. Although the lipid component has been impli-
cated in psoriasis development, lipidomic assays are not used
to extensively evaluate the efficacy of MTX in the treatment
of psoriasis. The aim of this study was to detect alterations in
lipid profile in psoriasis after MTX treatment.
Material and Methods
Study Design
From November 2017 to March 2018, 96 patients with psoriasis
were recruited from the Outpatient Department of Dermatology.
The first batch of samples were from 50 patients and the sec-
ond batch of samples were from 46 patients.
This study was conducted in accordance with the principles
of the World Medical Association Declaration of Helsinki, and
the research protocol was approved by the Medical Ethics
Committee of the First Affiliated Hospital of Anhui Medical
University (PJ2017-11-14). All patients provided informed con-
sent. Only patients over 18 years of age and with moderate-to-
severe psoriasis were included in this study. To minimize the
confounding factors, we excluded patients with classical phe-
notypes of hyperlipidemia and hyperglycemia. Patients with
psoriasis or arthropathy who received UV radiation, MTX, or
other systemic therapy within the first month of the study
were excluded. Additionally, topical therapy was discontinued
2 weeks before initiating MTX therapy. European guidelines
on contraindications and limitations of MTX use were strictly
followed. Psoriasis was assessed by 2 board-certified derma-
tologists at baseline and after 12 weeks using the Psoriasis
Area and Severity Index (PASI) and body surface area scores.
Patients were classified as responders if the PASI improved by
at least 75% from baseline after 12 weeks of treatment. Non-
responders showed a PASI improvement of less than 75%.
e943360-2
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

Lipid Extraction
Serum samples were taken from a -80°C freezer and thawed
at 4°C. Next, 1 mL of methanol/MTBE (1/3, v/v) and 20 µL of
internal lipid standard mix were added to 50 µL of the serum
samples. Each sample was vortexed for 3 min and sonicated
for 5 min; then, 500 µL of water was added. The supernatant
was separated using centrifugation at 12 000 rpm for 10 min
at 4°C and concentrated to obtain a powder. The powder was
redissolved in 100 μL of a mixture of isopropanol/acetonitrile
(1: 1) and analyzed using ultra-performance liquid chromatog-
raphy-mass chromatography.
Conditions for Liquid Chromatography-Mass Spectrometry
The treated sera were analyzed using liquid chromatography on
2.6 μm, 2.1×100 mm Thermo C30 columns (Thermo Scientific,
Waltham, MA, USA) using a ultra-performance liquid chroma-
tography system (Shim-pack UFLC CBM30A; Shimadzu, Kyoto,
Japan). The eluate was analyzed using mass chromatography
(QTRAP 6500; Applied Biosystems, Foster, CA, USA).
Mobile phase A was 0.04% (v/v) acetic acid in water, and mo-
bile phase B was 0.04% (v/v) acetic acid in acetonitrile; the
flow rate was 0.35 mL.min
-1
. Mobile phase gradient condi-
tions were as follows: A/B (80: 20, v/v) 0 min, 70: 30 v/v 2.0
min, 40: 60 v/v 4 min, 15: 85 v/v 9 min, 10: 90 v/v 14 min,
5: 95 v/v 15.5 min, 5: 95 v/v 17.3 min, 80: 20 v/v 17.3 min,
and 80: 20 v/v 20 min. The column and autosampler temper-
atures were maintained at 45°C and 4°C, respectively. The in-
jection volume was 2 μL. Instrument tuning and mass calibra-
tion were performed using 10 and 100 μmol·L
–1
polypropylene
glycol solutions. Scanning detection of ion pairs is based on
optimized cluster potential and collision energy. The chemi-
cal structures and contents of its metabolites were analyzed
using triple quadrupole scanning and multiple reaction mon-
itoring techniques.
Qualitative and Quantitative Analyses of Metabolites
Qualitative analysis of first and second mass spectrometry data
for candidate metabolites was performed using a self-created
MetWare database and a public database of metabolite infor-
mation. To ensure data accuracy, peak-to-peak integration cor-
rections were performed on mass spectra of different samples
of the same metabolite. The relative abundance of metabo-
lites was obtained by calculating the peak area of each chro-
matogram in the mass spectrometry data.
Statistical Analysis
Owing to the missing values of some samples and metab-
olites, the following procedure was used to obtain the final
dataset: (1) the metabolites with 40% missing values were
excluded; (2) the missing values were imputed with the R
“mice” package, using the default method “pmm” to pre-
dict the mean levels. The differentially expressed metabolites
were detected using the R “limma” package, with the multi-
ple linear regression model, “limFit”, with age, sex, and body
mass index as covariates. Results with P<0.05 were consid-
ered significant. Pathway enrichment analysis was conduct-
ed using the Kyoto Encyclopedia of Genes and Genomes da-
tabase (http://www.genome.jp/kegg/).
Results
Clinical Characteristics of the Recruited Patients
A total of 96 samples were analyzed, and the clinical charac-
teristics of the patients from whom the samples were collect-
ed, such as demographic analysis of sex, age, body mass index,
smoking habit, and alcohol consumption, are shown in Table 1.
The first batch samples were from 50 patients with psoriasis
(36 male patients, 72.0%) and the second batch samples were
from 46 patients with psoriasis (32 male patients, 70.0%).
Detailed information for batch 1 and 2 samples is presented
below. The mean age at the time of the survey was 42.7±11.4
and 43.5±12.8 years, respectively, and mean age of initial on-
set was 30.9±13.4 and 31.1±13.2 years, respectively. The du-
ration of psoriasis was 11.8±9.7 and 12.4±11.5 years, respec-
tively. The mean PASI of these patients with psoriasis was
21.2±14.0 and 12.4±6.8, respectively, at 0 weeks.
After 12 weeks of MTX treatment, 96 patients were divided
into 2 groups: MTX responders (35%) and MTX non-respond-
ers (65%). In batch 1, 10 (20%) of the 50 patients were re-
sponders, whereas 40 (80%) were non-responders. In batch 2,
24 (52%) of the 46 patients were responders, while 22 (48%)
were non-responders. There were no significant differences
between responders and non-responders in terms of age, sex,
body mass index, smoking habit, alcohol consumption, disease
progression, PASI score at 0 weeks, and MTX dose. Age of dis
-
ease onset was higher in responders than in non-responders,
although there was no significant difference. However, the 2
groups showed significant differences in PASI scores and PASI
improvement at 12 weeks after MTX treatment. Specifically,
for the response group, the PASI improvement was 83.8±6.3%
in batch 1 and 82.4±6.4% in batch 2, whereas for the non-re-
sponse group, the PASI improvement was 9.7±34.9% in Batch
1 and 8.0±38.8% in Batch 2 at 12 weeks (Table 1).
e943360-3
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

Lipidomic Assays in the Discovery and Validation Cohorts
As shown in the flow chart in Figure 1, 2 lipid assays were used
to determine the metabolite levels in the 2 cohorts. The raw lip-
id levels were detected using liquid chromatography-mass spec-
trometry and analyzed using SIMCA-P software. As shown in
Figure 1A and 1B, 717 lipids (64.8% TG, 15.5% PC, 7% cerami-
des, and 4% phospholipids (PE)) were detected in assay 1 and
1145 (26.2% TG, 19% PE, 7.5% carnitine, and 5.5% PC) were de-
tected in assay 2. The number of shared lipid metabolites was
316 (26.9% PC, 19.6% TG, 15.8% PE, 9.8% diglycerides (DG), and
9.5% LPC), and that of the total lipids was 1546 (34.8% TG, 14.5%
PE, and 12.4% PC) in the 2 assays (Figure 1C, 1D).
Lipid Metabolites Correlated with the Age of Onset
Owing to the obvious bimodal age of onset for psoriasis, we
suspected that lipid metabolites would be altered accord-
ing to age-related changes (Figure 2). On screening for all
metabolites, we found that some lipid metabolites significant-
ly correlated with age of onset. In cohort 1, DG (16: 1/18: 1)
and DG (18: 1/18: 1) positively correlated with age of onset:
P=2.00E-03, rPearson=0.43 and P=3.25E-03, rPearson=0.41, respec-
tively (Figure 2C). On classifying the patients according to EOP
(<40 years) and LOP (³40 years), we noted that DG (16: 1/18: 1)
and DG (18: 1/18: 1) clearly separated EOP and LOP, with area
under the receiver operating characteristic curve (AUC) scores
of 0.84 and 0.81, respectively. This finding was also verified
in cohort 2. DG (16: 1/18: 1) and DG (18: 1/18: 1) positively
correlated with the age of onset with P=1.54E-04, r
Pearson
=0.53
and P=8.20E-05, rPearson=0.55 (Figure 2D). The AUC scores were
0.76 and 0.77 for DG (16: 1/18: 1) and DG (18: 1/18: 1) in the
distinction between EOP and LOP.
MTX Treatment and Metabolic Alteration
MTX treatment-induced metabolic alteration has been dem-
onstrated in rheumatic diseases. To elucidate whether MTX
N Age Sex
(Male: Female) BMI Smoking
(%)
All samples
Batch 1 50 42.7±11.4 36: 14 24.1±3.7 23 (46%)
Batch 2 46 43.5±12.8 32: 14 24.4±3.0 17 (37.0%)
MTX Responser (RE)
Batch 1 10 46.2±9.6 6: 4 24.6±3.4 5 (50%)
Batch 2 24 46.0±12.3 16: 8 24.0±2.7 10 (41.7%)
MTX Non-responser (NR)
Batch 1 40 41.9±10.9 30: 10 23.9±3.8 18 (45%)
Batch 2 22 40.9±13.0 16: 6 24.9±3.3 7 (31.8%)
Table 1. Clinical characteristics of patients with psoriasis on methotrexate therapy.
PASI – psoriasis area and severity index; BMI – body mass index; RE – PASI improved by at least 75%; NR – PASI improved by less than
75%.
N Age of onset Duration
(ys)
PASI
(0w)
Dosage
(mg)
All samples
Batch 1 50 30.9±13.4 11.8±9.7 21.2±14.0 118.2±21.6
Batch 2 46 31.1±13.2 12.4±11.5 12.4±6.8 124.9±22.3
MTX Responser (RE)
Batch 1 10 28.9±20.3 4.4±2.9 83.8±6.3% 119.5±13.6
Batch 2 24 14.5±7.4 2.7±2.0 82.4±6.4% 116.5±15.6
MTX Non-responser (NR)
Batch 1 40 19.6±12.1 16.3±9.0 9.7±34.9% 117.8±23.3
Batch 2 22 10.1±5.5 8.9±5.6 8.0±38.8% 134.1 ±25.2
e943360-4
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

treatment induces metabolic alteration in psoriasis, we com-
pared the 1546 metabolic features between 0 and 12 weeks
of MTX treatment. Principal component analysis showed
remarkable separation between 0 and 12 weeks of MTX
treatment in cohort 1 (Figure 3A). At a false discovery rate-
adjusted P value of 0.01, we identified 314 differentially ex-
pressed targets (Figure 3B). Interestingly, all metabolites were
downregulated in the MTX treatment group, with 63.7% be-
ing the TG component. Furthermore, the top 3 TG targets
were TG (16: 0/18: 0/22: 0) (HMDB0043916, P=9.67E-07),
TG (18: 0/18: 0/22: 0) (HMDB0044751, P=1.23E-06), and TG
(14: 0/18: 0/22: 0) (HMDB0042156, P=1.23E-06) (Figure 3C).
Further pathway analysis revealed that the 139 differentially ex-
pressed lipids were enriched in glycerophospholipid (Figure 3D).
Metabolic Alteration in Responders and Non-Responders of
MTX Treatment
The MTX responders accounted only for 35% of all patients
with psoriasis. Whether metabolic alterations led to the effica-
cy of MTX treatment was clearly demonstrated. Therefore, we
performed differential lipid analysis in responders and non-re-
sponders. The principal component analysis showed an obvious
separation between the groups (Figure 4A). The volcano plot
indicated 5 differentially expressed lipids, including 3 upreg-
ulated and 2 downregulated targets (Figure 4B). The 3 upreg
-
ulated lipids consisted of 2 PCs, PC (O-30: 2) (HMDB0013410,
P=2.58E-03) and PC (O-36: 1) (HMDB0013427, P=4.46E-03),
and a ceramide, ceramide P (d18: 1/16: 0), (HMDB0010700,
P=4.78E-03). The 2 downregulated lipids were phosphatidyl-
ethanolamine PE (34: 4) (HMDB0008838, P=8.61E-03) and PE
(38: 1) (HMDB0008849, P=9.42E-03) (Figure 4C). The pathway
analysis showed that these 5 lipid components were highly
64.8%
Metabolite 1
15.5%
1.2%
1.2%
1.5%
1.7% 3%
4%
7% TG
PC
Cer
PE
Others
DG
PS
CerP
LPE
26.2%
26.2%
Metabolite 2
7.5%
19%
3% 3.3% 4%
4.9%
5%
5.5%
TG
Others
PE
CAR
PC
SM
PI
FFA
LPC
HexCer
26.9%
19.6%
Common
9.5%
9.8%
15.8%
1.9% 2.5%
3.5%
4.7%
5.7% PC
TG
PE
DG
LPC
LPE
CE
Cer
Others
HexCer
20.2%
34.8%
All
14.5%
2.4% 2.5% 2.6%
2.7%
3.7%
4%
12.4%
TG
Others
PE
PC
CAR
LPC
SM
PI
Cer
DG
A
C
B
D
Figure 1. Distribution of lipid metabolites in the 2 assays. (A, B) A total of 717 and 1145 lipids were identified in assays 1 and 2,
respectively. (C, D). The shared 316 lipid metabolites and a total of 1,546 lipids in the 2 assays are shown in the pie plot.
(Figures were generated by R with ggplot2_3.4.2.).
e943360-5
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

0.6
0.4
0.2
0.0
-0.2
-0.4
DG metabolites in cohort 1
Pearson correlation coecient
0.6
0.4
0.2
0.0
-0.2
-0.4
DG metabolites in cohort 2
17
16
15
14
20
Age onset
40
P=2.00e-03, rPearson=0.43
P=3.25e-03, rPearson=0.41
P=1.54e-04, rPearson=0.53
P=8.20e-05, rPearson=0.55
60
DG(16:1/18:1)
21.0
20.5
20.0
19.5
20
Age onset
Cohort 1
40 60
DG(18:1/18:1)
19
17
15
2010 30
Age onset
40 50
2010 30 40 50
24
23
22
21
20
Age onset
Cohort 2
Pearson correlation coecient
A
B
C D
Figure 2. Lipid metabolites correlated with the age of onset in patients with psoriasis. (A, B) The distribution of Pearson correlation
coefficients for diglyceride (DG) metabolites in cohorts 1 and 2. Each dot represents a DG component. The component with
correlation coefficient >0.3 or <-0.3, and correlation P value <0.05 are marked. (C) The change in the level of DG (16: 1/18: 1)
and DG (18: 1/18: 1) with the age of onset in cohorts 1. (D) The change in the level of (16: 1/18: 1) and DG (18: 1/18: 1) with
the age of onset in cohorts 2. (Figures were generated by R with ggplot2_3.4.2.).
e943360-6
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

7.5
5.0
2.5
0.0
-2.5
11
11
11
10
10
10
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.00 0.05 0.10
Pathway impact
Glycerophospholipid metabolism
0.200.15
-log10(p)
-10 -5 -1 -0.5 0.0
Fold change Total=564 variables
0.5 1.00510
MTX treatment of psoriasis
X-variate 2: 4% expl. var
7.5
5.0
2.5
0.0
-Log10 P
X-variate 1: 30% expl. var
TG(16:0/18:0/22:0)(log2(intensity))
Legend
0W
12W
0W
p=9.67e-07
12W
10
10
10
TG(18:0/18:0/22:0)(log2(intensity))
0W
p=1.23e-06
12W
11
10
10
TG(14:0/18:0/22:0)(log2(intensity))
0W
p=1.23e-06
12W
11
11
11
10
10
10
TG(14:0/20:0/20:0)(log2(intensity))
0W
p=1.32e-06
12W
A B
C
enriched in glycerophospholipids, and the same pathway was
identified on MTX treatment (Figure 4D).
Discussion
The pathogenesis of psoriasis has not been elucidated.
Identifying biomarkers for psoriasis diagnosis and treatment
response remains a challenge, although some genomic, tran-
scriptomic, proteomic, and metabolomic biomarkers have been
identified in the past few decades [12]. More than 20% of pa-
tients with psoriasis have a metabolic syndrome, suggesting
that aberrant metabolic alterations can provide some clues
for disease treatment.
Lipidomics provides a systematic perspective on lipid altera-
tions caused by environmental, genetic, and multiple organ in-
teractions. The lipids associated with psoriasis are primarily
found in the stratum corneum, skin surface, and serum, espe-
cially TGs or ceramides [7,13,14]. The total cholesterol and TG
levels, along with low-density lipoproteins (LDL) and very LDLs,
in the blood of patients with psoriasis are significantly higher,
whereas high-density lipoprotein concentration is significant-
ly lower than that in healthy individuals [15]. The more severe
e943360-7
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

7.5
5.0
2.5
0.0
-2.5
11
11
11
10
10
10
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.00 0.05 0.10
Pathway impact
Glycerophospholipid metabolism
0.200.15
-log10(p)
-10 -5 -1 -0.5 0.0
Fold change Total=564 variables
0.5 1.00510
MTX treatment of psoriasis
X-variate 2: 4% expl. var
7.5
5.0
2.5
0.0
-Log10 P
X-variate 1: 30% expl. var
TG(16:0/18:0/22:0)(log2(intensity))
Legend
0W
12W
0W
p=9.67e-07
12W
10
10
10
TG(18:0/18:0/22:0)(log2(intensity))
0W
p=1.23e-06
12W
11
10
10
TG(14:0/18:0/22:0)(log2(intensity))
0W
p=1.23e-06
12W
11
11
11
10
10
10
TG(14:0/20:0/20:0)(log2(intensity))
0W
p=1.32e-06
12W
D
Figure 3. Lipid metabolic alterations in patients with psoriasis after methotrexate (MTX) treatment. (A) Principal component
analysis plot clearly separating samples from 0 and 12 weeks of MTX treatment in cohort 1. (B) Volcano plot showing
139 differentially expressed metabolites. (C) Three representative lipids were significantly altered after 12 weeks of MTX
treatment. Each dot indicates an individual. The 1 quantile and mean and 3 quantile levels are shown in the boxplot.
(D) Metabolic pathways showed that the differentially expressed lipids are prominent in the glycerophospholipid pathway.
(Figures were generated by R with ggplot2_3.4.2, EnhancedVolcano_1.16.0. The metabolic pathways analysis was based on
MetaboAnalyst 6.0.).
the psoriasis, the more obvious the abnormal changes in lip-
id metabolism. It is speculated that high levels of TG are pos-
itively associated with an increased risk of psoriasis [14,16].
Moreover, the levels of other lipid metabolites, such as free
fatty acids, cholesterol, and PE, were abnormal in the serum
of patients with psoriasis per the findings of previous studies.
To evaluate the levels of serum lipids in patients with psoria-
sis with different phenotypes and the change after MTX treat-
ment, we first characterized the serum lipid profiles of 96 pa-
tients with psoriasis and further identified the differentially
expressed lipids between EOP and LOP. Some previous stud-
ies have shown that EOP and LOP differ in their clinical char-
acteristics and genetic background [5,17,18]. By comparing
the serum lipid metabolites in LOP and EOP, we found that
the levels of DG (16: 1/18: 1) and DG (18: 1/18: 1) significant-
ly increased in EOP. This is the first attempt to compare lip-
id metabolism abnormalities between the psoriasis types. In
normal human plasma, DGs were also present but at substan-
tially lower levels [19]. DGs are important lipid mediators and
can stimulate the activity of protein kinase C, thus acting as
membrane second messengers. The activation of DG-signal
transduction pathways can regulate IL-6 expression. Injured
skin is one of the major sites of IL-6 production, and elevated
levels of IL-6 are observed in individuals with psoriasis [20].
Systemic drugs used in the treatment of psoriasis plaques
also affect serum lipid levels. Acitretin can cause hyperlipid-
emia as an adverse effect. Biological products, such as anti-
IL-17A, anti-TNFa, and anti-IL-12/23 antibodies, affect the se-
rum lipid metabolism in patients to different degrees before
and after treatment.
MTX has anti-inflammatory, antiproliferation, and immunosup-
pressive effects and is widely employed as a treatment for in-
flammatory disorders and autoimmune diseases, such as pso-
riasis, lupus erythematous, dermatomyositis, and rheumatoid
arthritis. However, our understanding of the mechanism of
psoriasis treatment with low-dose MTX is limited. The mech-
anism of action of MTX in the treatment of psoriasis is main-
ly mediated by activated T-cell death, and highly activated T
cells are more susceptible to the cytotoxic effects induced by
MTX, which can cause a decrease in the levels of cytokines,
such as IL-1, IL-2, IL-4, IL-8, INF-g, and TNF-a. Some patients
with psoriasis respond well to MTX, whereas others respond
poorly. Some studies reported that IL10 was significantly in-
creased after 12 weeks of MTX treatment [21]. Therefore, we
attempted to analyze the mechanism of action of MTX in pso-
riasis from the perspective of lipid metabolism. This is the first
study to assess the effect of MTX on lipid metabolism in pa-
tients with psoriasis, although serum biomarkers have been
e943360-8
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

identified to evaluate the efficacy of MTX treatment in patients
with early rheumatoid arthritis [22].
Phospholipids, namely PE, also known as cephalin, accounts
for 15% to 25% of the total phospholipids (major components
of biomembranes) in cells, and is the second most abundant
phospholipid in mammalian cells [23]. PEs are involved in the
regulation of cell proliferation, metabolism, organelle function,
endocytosis, autophagy, stress responses, apoptosis, and ag-
ing [24]. The concentrations of PCs, essential parts of the cell
membrane, are low in the serum of patients with psoriasis,
and this is mostly related to the higher rate of proliferation
of skin cells [25]. LPC is formed by the cleavage of PC and can
be divided into saturated and unsaturated LPC according to
the acyl chain length and degree of saturation of LPC. An in-
crease or decrease in LPC levels can be observed in different
diseases, such as cardiovascular diseases, inflammatory dis-
eases, diabetes, and tumors [26]. LPC (16: 0) and LPC (18: 1)
exert pro-inflammatory effects by increasing the release of
inflammatory cytokines, such as IL-6 and IL-8 [27]. LPC is the
20
10
0
-10
-20
10.2
10.0
9.8
9.6
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.00 0.05 0.10
Pathway impact
Glycerophospholipid metabolism
0.20 0.250.15
-log10(p)
-10-15 -5
-0.1 0.0
Fold change
0.1
0510
MTX responser vs nonresponser
X-variate 2: 12% expl. var
3
2
1
0
-Log10 P
X-variate 1: 28% expl. var
PE(34:4)log2(intensity))
Legend
NR
RE
NR_12W RE_12W
p=0.008
10.55
10.50
10.45
10.40
10.35
10.30
PE(38:1)log2(intensity))
NR_12W RE_12W
p=0.009
A
C
B
D
Figure 4. Metabolic alterations among responders and non-responders of methotrexate (MTX) treatment. (A) Principal component
analysis showed obvious separation between the responders and non-responders. (B) Volcano plot indicates 5 differentially
expressed lipids, including 3 upregulated and 2 downregulated targets. (C) Phosphatidylethanolamine PE (34: 4) and PE
(38: 1) are downregulated in the responder group. (D) Five lipid components are highly enriched in the glycerophospholipid
pathway. (Figures were generated by R with ggplot2_3.4.2, EnhancedVolcano_1.16.0. The Metabolic pathways analysis was
based on MetaboAnalyst 6.0.).
e943360-9
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

main component of oxidatively damaged LDL, and numerous
studies have reported that patients with psoriasis have elevat-
ed oxidatively damaged LDL levels. A previous study showed
that the lysophosphatidic acid, LPC, and PA levels significantly
increase, whereas the phosphatidylinositol and PC levels de-
creases in the plasma of patients with psoriasis [11].
In the present study, we detected significant changes in serum
lipid levels. The levels of TG (16: 0/18: 0/22: 0), TG (18: 0/18:
0/22: 0), and TG (14: 0/18: 0/22: 0) decreased after 12 weeks
of MTX treatment. We concluded that MTX can inhibit inflam-
mation and attenuate psoriasis symptoms by reducing TG lev-
els. MTX treatment would alter the composition of serum lip-
ids. A recent study revealed that MTX decreased the levels of
pro-atherogenic lipids and the ApoB/ApoA1 ratio in male pa-
tients with psoriasis [28]. Meanwhile, genetic variations have
been reported to be associated with MTX response. In AnxA6,
the rs11960458 CC carriers obtained the lower levels of pro-
atherogenic lipids TC, LDL, and ApoB, compared with TC and TT
carriers [29], highlighting the genetic predisposition for the MTX
treatment mainly through the regulation of lipid metabolites.
Psoriasis is a chronic inflammatory disease, and inflamma-
tion can induce a variety of alterations in lipid metabolism.
During inflammation and infection, various cytokines are pro-
duced, such as TNF, IL-1, and IL-6, which alter the body’s lipid
metabolism. In our study, we found that the PC (O-30: 2), PC
(O-36: 1), and ceramide P levels increased, while PE (34: 4) and
PE (38: 1) levels decreased in MTX responders. The increased
serum levels of PC in responders may be related to the decline
in the proliferation of skin cells. These changes in PE concen-
trations may be related to the therapeutic effects of MTX, war-
ranting further in-depth study. This would help us to evaluate
the efficacy of MTX, from the perspective of lipid metabolism.
For those MTX non-responders, it would be very important to
add some blood lipid regulators combined with MTX therapy
or use some other immune-suppressant drug.
Some studies have shown that dysregulated metabolic path-
ways are related to the pathogenesis of psoriasis. Amino acid
metabolic activity and glycolysis are upregulated, whereas the
fatty acid biosynthesis pathway is downregulated in patients
with psoriasis [30]. A study of serum lipid metabolism in 75
patients with psoriasis and 75 healthy controls yielded 44 po-
tential serum biomarkers, which are prominent in the path-
ways of glycerophospholipid metabolism, sphingolipid metab-
olism, arachidonic acid metabolism, and bile acid biosynthesis.
In patients with psoriasis, the components of glycerophos-
pholipid metabolism are disordered. In the present study, the
highly expressed lipids were enriched in glycerophospholip-
id metabolism.
Our study has certain limitations that warrant consideration.
First, we recruited only 96 patients, which is not sufficient to
represent the different types of patients. Further studies with
a higher number of patients are required to further evaluate
the effect of lipids on psoriasis. Second, the dynamics and sta-
bility of lipid profiles need to be established for better clinical
applications in the future. Third, we only examined changes
in lipid profiles in Chinese patients with psoriasis after MTX
treatment. It is not clear whether the same changes occur in
patients from other ethnicities. Fourth, lipid metabolites would
alter according to the individual lifestyles, this detailed infor-
mation was missed in our study.
Conclusions
We employed serum lipidomics to investigate the potential
molecular and physiological mechanisms of MTX treatment
in psoriasis. These lipids can be potential targets for psoriasis
treatment in the future. Nevertheless, the relationship between
these lipid profile alterations and the effect of MTX treatment
for psoriasis remains to be investigated.
Acknowledgements
The authors thank all the participants of the study.
Declaration of Figures’ Authenticity
All figures submitted have been created by the authors, who
confirm that the images are original with no duplication and
have not been previously published in whole or in part.
References:
1. Roberson ED, Bowcock AM. Psoriasis genetics: Breaking the barrier. Trends
Genet. 2010;26(9):415-23
2. Ding X, Wang T, Shen Y, et al. Prevalence of psoriasis in China: A popula-
tion-based study in six cities. Eur J Dermatol. 2012;22(5):663-67
3. Griffiths CEM, Armstrong AW, Gudjonsson JE, Barker J. Psoriasis. Lancet.
2021;397(10281):1301-15
4. Gupta R, Debbaneh MG, Liao W. Genetic epidemiology of psoriasis. Curr
Dermatol Rep. 2014;3(1):61-78
5. Huseynova Terzi L, Dogan Gunaydin S. Comparison of early and late
onset psoriasis (EOP and LOP) regarding systemic inflammatory comorbid-
ities: LOP is a more rapid subtype of psoriasis. Dermatol Pract Concept.
2022;12(3):e2022144
6. Hao Y, Zhu YJ, Zou S, et al. Metabolic syndrome and psoriasis: Mechanisms
and future directions. Front Immunol. 2021;12:711060
7. Nowowiejska J, Baran A, Flisiak I. Aberrations in lipid expression and me-
tabolism in psoriasis. Int J Mol Sci. 2021;22(12):6561
8. Alqarni AM, Zeidler MP. How does methotrexate work? Biochem Soc Trans.
2020;48(2):559-67
e943360-10
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)

9. Yan K, Zhang Y, Han L, et al. Safety and efficacy of methotrexate for Chinese
adults with psoriasis with and without psoriatic arthritis. JAMA Dermatol.
2019;155(3):327-34
10. Yan KX, Zhang YJ, Han L, et al. TT genotype of rs10036748 in TNIP1 shows
better response to methotrexate in a Chinese population: A prospective
cohort study. Br J Dermatol. 2019;181(4):778-85
11. Zeng C, Wen B, Hou G, et al. Lipidomics profiling reveals the role of glyc-
erophospholipid metabolism in psoriasis. Gigascience. 2017;6(10):1-11
12. Jiang S, Hinchliffe TE, Wu T. Biomarkers of an autoimmune skin disease –
psoriasis. Genomics Proteomics Bioinformatics. 2015;13(4):224-33
13. Pietrzak A, Michalak-Stoma A, Chodorowska G, Szepietowski JC. Lipid
disturbances in psoriasis: An update. Mediators of inflammation.
2010;2010:535612
14. Dreiher J, Weitzman D, Davidovici B, et al. Psoriasis and dyslipidaemia: A
population-based study. Acta Derm Venereol. 2008;88(6):561-65
15. Pietrzak A, Chabros P, Grywalska E, et al. Serum lipid metabolism in psori-
asis and psoriatic arthritis – an update. Arch Med Sci. 2019;15(2):369-75
16. Snekvik I, Nilsen TIL, Romundstad PR, Saunes M. Metabolic syndrome and
risk of incident psoriasis: Prospective data from the HUNT Study, Norway.
Br J Dermatol. 2019;180(1):94-99
17. Chularojanamontri L, Kulthanan K, Suthipinittharm P, et al. Clinical differenc-
es between early- and late-onset psoriasis in Thai patients. Int J Dermatol.
2015;54(3):290-94
18. Zhou Y, Cai M, Sheng Y, Zhang X. A large-scale, stratified genetic analysis of
the major histocompatibility complex region in early- and late-onset pso-
riasis in China. Ann Dermatol. 2021;33(1):61-67
19. Quehenberger O, Armando AM, Brown AH, et al. Lipidomics reveals a remark-
able diversity of lipids in human plasma. J Lipid Res. 2010;51(11):3299-305
20. Sehgal PB. Interleukin-6: Molecular pathophysiology. J Invest Dermatol.
1990;94(6 Suppl.):2S-6S
21. Owczarczyk-Saczonek A, Drozdowski M, Maciejewska-Radomska A, et al.
The effect of subcutaneous methotrexate on markers of metabolic syn-
drome in psoriatic patients – preliminary report. Postepy Dermatol Alergol.
2018;35(1):53-59
22. Wang Z, Chen Z, Yang S, et al. (1)H NMR-based metabolomic analysis for
identifying serum biomarkers to evaluate methotrexate treatment in pa-
tients with early rheumatoid arthritis. Exp Ther Med. 2012;4(1):165-71
23. Vance JE. Phospholipid synthesis and transport in mammalian cells. Traffic
(Copenhagen, Denmark). 2015;16(1):1-18
24. Patel D, Witt SN. Ethanolamine and phosphatidylethanolamine: Partners
in health and disease. Oxid Med Cell Longev. 2017;2017:4829180
25. Ottas A, Fishman D, Okas TL, et al. The metabolic analysis of psoriasis iden-
tifies the associated metabolites while providing computational models for
the monitoring of the disease. Arch Dermatol Res. 2017;309(7):519-28
26. Liu P, Zhu W, Chen C, et al. The mechanisms of lysophosphatidylcholine in
the development of diseases. Life Sci. 2020;247:117443
27. Chang MC, Lee JJ, Chen YJ, et al. Lysophosphatidylcholine induces cytotox-
icity/apoptosis and IL-8 production of human endothelial cells: Related
mechanisms. Oncotarget. 2017;8(63):106177-89
28. Wang B, Deng H, Hu Y, et al. The difference of lipid profiles between pso-
riasis with arthritis and psoriasis without arthritis and sex-specific down-
regulation of methotrexate on the apolipoprotein B/apolipoprotein A-1 ra-
tio. Arthritis Res Ther. 2022;24(1):17
29. Zhang F, Han L, Wang B, et al. Annexin A6 polymorphism is associated with
pro-atherogenic lipid profiles and with the downregulation of methotrexate
on anti-atherogenic lipid profiles in psoriasis. J Clin Med. 2022;11(23):7059
30. Kang H, Li X, Zhou Q, et al. Exploration of candidate biomarkers for human
psoriasis based on gas chromatography-mass spectrometry serum metab-
olomics. Br J Dermatol. 2017;176(3):713-22
e943360-11
Indexed in: [Current Contents/Clinical Medicine] [SCI Expanded] [ISI Alerting System]
[ISI Journals Master List] [Index Medicus/MEDLINE] [EMBASE/Excerpta Medica]
[Chemical Abstracts/CAS]
Zhou Y. et al:
Lipid profiles alterations with methotrexate
© Med Sci Monit, 2024; 30: e943360
CLINICAL RESEARCH
This work is licensed under Creative Common Attribution-
NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0)