LncRNA Profile in Malaysian -thalassaemia (Mohd Yasin N, et al.
Indones Biomed J.
: 436-46
DOI: 10.
18585/inabj.
RESEARCH ARTICLE
Long Noncoding RNAs TYMSOS.
VASH1-AS1, and LINC01001 Expressions as Biomarkers of -thalassaemia Severity Among the Malaysian Patients Norafiza Mohd Yasin1.
Nur Fatin Aqilah Raman1.
Siti Aishah Sulaiman2,E.
Nurhairulamri Kosnan1.
Nor Azian Abdul Murad2.
Ermi Neiza Mohd Sahid1.
Yuslina Yusoff1.
Hishamshah Mohd Ibrahim3.
Zulaiha Muda3.
Ezalia Esa1 Haematology Unit.
Cancer Research Centre.
Institute for Medical Research (IMR).
National Institutes for Health.
Persiaran Setia Murni.
Setia Alam, 40170 Shah Alam.
Selangor.
Malaysia UKM Medical Molecular Biology Institute (UMBI).
Universiti Kebangsaan Malaysia.
Jalan Yacob Latif.
Bandar Tun Razak, 56000 Cheras.
Wilayah Persekutuan Kuala Lumpur.
Malaysia Paediatric Department.
Women and Children Hospital.
Tunku Azizah Hospital (HTA).
Lot 25.
Jalan Raja Muda Abdul Aziz.
Kampung Baru, 50300 Kuala Lumpur.
Malaysia *Corresponding author.
Email: sitiaishahsulaiman@ukm.
Received date: Jul 21, 2025.
Revised date: Sep 2, 2025.
Accepted date: Sep 11, 2025 Abstract ACKGROUND: -thalassaemia has heterogeneous disease severities ranging from mild to trait, and major.
Long noncoding RNAs .
ncRNA.
are known to regulate microRNAs .
iRNA.
and genes, possibly modifying the disease phenotypes.
However, limited data exist on the lncRNAs in -thalassaemia, and no study has yet addressed this gap in Malaysia.
Therefore, this study aimed to identify the expression profile of lncRNAs in -thalassaemia major and trait among Malaysians.
METHODS: Case-control study was conducted in two phases at Tunku Azizah Women and Children's Hospital and the Institute of Medical Research, from September 2019 to November 2021.
Total of 141 individuals were recruited, comprising -thalassemia major (MAJOR, n=.
, -thalassemia trait (TRAIT, n=.
, and healthy controls (CON, n=.
All participants were genotyped for thalassaemia and assessed for their haemoglobin and red blood cells (RBC) indices.
In the first phase, discovery of lncRNA was performed using microarray, and differential expression of lncRNAs (DEL) in MAJOR.
TRAIT, and CON groups was identified.
Significant lncRNAs were subjected to lncRNA-miRNA prediction, and gene ontology and biological pathway were analyzed.
In validation phase, six potential lncRNAs were further validated via using lncRNA polymerase chain reaction (PCR) custom array.
RESULTS: Total of 364 DELs were identified in MAJOR group, and 128 DELs were identified in TRAIT group.
Between the MAJOR and TRAIT groups, 100 DELs were dysregulated in MAJOR group.
Two molecular networks comprising the lncRNA interactions with miRNAs were identified and associated with traits and major phenotypes, resulting in six potential lncRNAs for validation.
Among these six lncRNAs, three lncRNAs (TYMSOS.
VASH1-AS1, and LINC01.
were reduced in the MAJOR group .
old change (FC)=-6.
67, p=0.
FC=-8.
33, p=0.
and FC=-8.
33, p=0.
021, respectivel.
CONCLUSION: Expressions of TYMSOS.
VASH1-AS1, and LINC01001 lncRNAs were altered differently between -thalassaemia major and trait patients.
Therefore these lncRNAs may serve as novel biomarkers for -thalassaemia disease severity in Malaysian population.
KEYWORDS: -thalassaemia, lncRNAs, miRNAs, severity, major, trait, molecular network Indones Biomed J.
: 436-46
Copyright A 2025 The Prodia Education and Research Institute.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.
0 International (CC-BY-NC) License.
The Indonesian Biomedical Journal.
Vol.
No.
October 2025, p.
Introduction -thalassaemia is an autosomal recessive disorder caused by mutations in the -globin (HBB) gene, resulting in abnormal haemoglobin production, ineffective erythropoiesis, and a shortened lifespan of red blood cells (RBC).
Until now, more than 950 mutations have been reported within the HBB gene and its flanking regions.
In Malaysia, -thalassaemia is among the most prevalent hereditary diseases .
% carrier.
and a major public health problem.
Therefore, significant efforts have been made to identify the molecular changes within the HBB gene among Malaysians in order to reduce the burden and improve patient outcomes.
Previous studies have identified mutations associated with -thalassaemia patients in Malaysia.
Among them, the most common mutations in Malaysian Malays are Hb Malay .
odon 19 (A>G) HBB:c.
59A>G.
IVS-I-1
(G>T) HBB:c.
92 1G>T.
IVS-I-5 (G>C) HBB:c.
92 5G>C,
polyadenylation signal .
olyA) (AATa>AATAGA) HBB:c.
*112A>G, and Hb E thalassaemia, (HBB:c.
79G>A).
On the other hand.
Malaysian Chinese patients are often presented with codons 41/42 (AeTTCT) HBB:c.
126_129delCt.
IVS-II-654 (C>T) HBB:c.
316197C>T, -28 (A>G) HBB:c.
-78A>G, and codon 17 (A>T) HBB:c.
52A>T.
The overall data also showed that the most commonly known form of -thalassaemia in Malaysia is the combination of Hb E and -thal (Hb E/tha.
, which accounted for 34.
37% of the patients .
, and is quite common among South-East Asians .
However, the clinical manifestations of -thalassaemia are highly variable, ranging from the mild thalassaemia trait to severe thalassaemia major.
This phenotype heterogeneity poses significant challenges in prevention strategies, prenatal screening, clinical diagnosis, and treatment planning.
Various factors contributed to genotype-phenotype heterogeneity in -thalassaemia.
The degree of chain imbalance and the mechanism of -thalassaemia mutations are primary factors in determining the -thalassemia .
,8,.
Due to these complex genetic factors and differential phenotypes, many studies investigated the additional genetic modifiers of -thalassaemia.
One such example is the long noncoding RNAs .
ncRNA.
, which bind to microRNAs .
iRNA.
and act as sponges to prevent them from exerting their function.
This lncRNA-miRNAgene network, also known as competitive endogenous RNA .
eRNA), which is a system of RNA interactions and co-regulatory networks, where both lncRNAs and genes compete for binding to specific miRNAs, resulting Print ISSN: 2085-3297.
Online ISSN: 2355-9179 in an increase or decrease in gene expression.
One of the earliest studies of lncRNAs in -thalassaemia reported that 605 lncRNAs were upregulated, and 257 lncRNAs were downregulated in -thalassemia minor compared to .
In the previous study, one miRNA, miR-4863p .
well-known regulator of -globin expression .
, was found to be associated with six lncRNAs in a negative Another study compared -thalassemia major, -thalassemia minor, and healthy controls and found that more than 3,000 lncRNAs were dysregulated in -thalassemia major patients, and over 1,000 lncRNAs were dysregulated in -thalassemia minor patients.
Among these dysregulated lncRNAs, lncRNA XIST, lincRNATPM1, and lincRNA-RUNX2-2 were upregulated, and lncRNA DQ583499 and MRFS16P were downregulated in -thalassemia individuals.
These findings suggest that lncRNAs could play a role in determining the severity of -thalassemia.
Unfortunately, none of these studies were performed on Malaysian patients.
Therefore, this pilot study aimed to determine the lncRNA expression profile and the potential molecular network of lncRNAs and miRNAs in Malaysian -thalassemia patients, whose preliminary clinical and genotypic data had been published previously.
Identification of these lncRNAs and their potential roles in determining disease severity may help unravel the underlying molecular landscape of -thalassemia and its associated phenotypes, thereby enabling the discovery of new biomarkers for disease severity assessment.
Methods Study Design and Participants This case-control study was a pilot study designed for two phases of investigation, and was conducted at the Tunku Azizah Women and Children's Hospital and the Institute of Medical Research.
Malaysia, from September, 2019 to November, 2021.
A total of 141 individuals were recruited, including -thalassemia major (MAJOR, n=.
, -thalassemia trait (TRAIT, n=.
, and healthy controls (CON, n=.
All participants were genotyped for thalassaemia, and details on their clinical phenotypes, including age, haemoglobin (H.
analysis, transfusion requirements, spleen size, and extramedullary changes, were collected from their clinical records.
The patients were considered -thalassemia major when they were transfusion-dependent .
equiring more than six transfusions yearl.
, had the 0 mutation, and had an enlarged spleen.
Whereas, -thalassemia trait patients were diagnosed based DOI: 10.
18585/inabj.
on Hb analysis and their -globin mutations.
Patients with other systemic diseases, such as liver disorders .
, chronic renal failure, hypothyroidism, and hyperthyroidism, were excluded from this study.
Healthy controls are included in the study if they have a normal complete blood count (CBC), normal Hb, and no - and -thalassemia genotypes, as well as no other existing genetic diseases and a family history of thalassemia.
This study was conducted in accordance with the guidelines of the Declaration of Helsinki and was approved by the Medical Research and Ethics Committee of the IMR (Ethics No.
: NMRR-192280-48428, dated September 20, 2.
Written informed consent was obtained from all subjects involved in the study.
The first phase of this study was the lncRNA discovery phase that included a total of nine participants comprising the healthy controls (CON, n=.
, -thalassemia major with homozygous Codon 41/42 (-TTCT) mutation (MAJOR, n=.
, and -thalassemia trait patients with heterozygous Codon 41/42 (-TTCT) mutation (TRAIT, n=.
from the all 141 participants.
The second phase involved validating the significant lncRNAs identified in the first phase in the remaining participants, comprising the MAJOR .
TRAIT .
, and CON .
Analysis of Blood Parameters All participants underwent analysis of their CBC and Hb levels.
The CBC was measured via the automated haematology analyser.
The Hb analysis was performed via the peripheral blood film, capillary electrophoresis (SEBIA.
Lisses.
Franc.
, via the HPLX (Bio-Rad Laboratories.
Hercules.
CA.
USA), and Hb electrophoresis (SEBIA), according to the standard protocol for thalassemia diagnosis in our centre.
The cut-off values for mean corpuscular haemoglobin (MCH) and mean corpuscular volume (MCV) were <25 pg and <80 fL, respectively.
DNA and RNA Isolation The whole blood was used to isolate DNA for thalassemia The DNA was isolated using the QIAamp DNA Blood Mini Kit (Qiagen Biotechnology.
Hilden.
German.
as recommended by the manufacturer.
The thalassemia genotyping was done as previously described .
as part of the standard protocol in thalassemia diagnosis.
The total RNA was isolated from whole blood samples using miRNeasy Serum/Plasma Kit (Qiagen Biotechnolog.
according to the manufacturer's protocol.
The RNA quality and RNA integrity number (RIN) were measured using the Agilent RNA 6000 Pico kit on the Agilent 2100 Bioanalyzer System (Agilent Technologies.
Santa Clara.
CA.
USA).
LncRNA Profile in Malaysian -thalassaemia (Mohd Yasin N, et al.
Indones Biomed J.
: 436-46
Only RNA samples with a RIN score of Ou8 were used for the downstream analysis.
Differential lncRNA Expression The discovery phase measured the expression of lncRNAs via the Agilent SurePrint G3 Human Gene Expression v3 8x60K microarray kit (Agilent Technologie.
as recommended by the manufacturer.
Raw data were extracted using the Agilent Feature Extraction Software (Agilent Technologie.
, and the raw data were normalised using the quantile normalization method via GeneSpring 14 (Agilent Technologie.
, as recommended by the The differential expression analysis was performed using the normalized expression data via the online analysis tool ExAtlas .
ttp://lgsun.
Three comparison analyses were performed: .
MAJOR vs.
CON, .
TRAIT vs.
CON, and .
MAJOR vs.
TRAIT groups.
Differentially expressed lncRNAs (DEL.
were selected based on the cut-off values of fold change (FC) >4 and adjusted p-value (Benjamin-Hochberg false discovery rat.
<0.
Prediction lncRNA-miRNAs Relationship All DELs were confirmed with the official lncRNA identification based on two lncRNA databases.
LNCipedia .
and NONCODE .
Using lncRNA-miRNA database LncBook v2 .
and miRNet .
, the predicted miRNA targets of each significant DELs were determined.
These predicted miRNAs were then subjected to a Venn diagram analysis via the EVenn tool .
to identify the commonly predicted miRNAs among the DELs.
The final list of predicted miRNAs was subjected to Gene Ontology (GO) enrichment and biological pathway analyses using g:Profiler .
, based on three known databases: the Kyoto Encyclopedia of Genes and Genomes (KEGG).
REACTOME, and WikiPathways.
An adjusted p<0.
05 was used to select the most significant GO and biological pathways.
Construction of the lncRNA-miRNA Network The list of DELs and associated predicted mRNAs was also used to identify the co-regulatory network between the molecules via miRNet .
and visualized using Cytoscape 3 .
The cytoHubba plugin was used to calculate the degrees, betweenness, and closeness centralities of the constructed networks.
Validation of the Significant DELs The expression of selected DELs was confirmed using Real-Time PCR array panel via the QuantiNova LNA The Indonesian Biomedical Journal.
Vol.
No.
October 2025, p.
lncRNA PCR custom panel (Catalogue No.
: SBCA96110 &
SBCA96404 Kit.
Qiagen Biotechnology.
Hilden.
German.
on a StepOne Plus real-time PCR system (Applied Biosystems.
Massachusetts.
USA), as recommended by the manufacturer.
Briefly, the custom array plate was run at the recommended cycle conditions .
AC of initial heat activation for 2 min, followed by 45 cycles of two-step cycling .
AC of denaturation for 5 seconds and 60AC of annealing/extension for 10 second.
The raw data were analysed via the GeneGlobe data analysis tool .
ttps:// com/us/), last accessed on January 10.
The differential expression was determined using the 2-iiCT method, and the p-value significance was set at <0.
05 for specific group comparisons, as recommended by the manufacturer.
The Glyceraldehyde 3-phosphate dehydrogenase (GAPDH).
Beta-actin (ACTB), and Beta2 microglobulin (B2M) genes were used as endogenous reference genes.
Statistical Analysis Statistical analysis was performed using SPSS Statistics 0 (IBM Corporation.
Armonk.
New York.
USA) using Print ISSN: 2085-3297.
Online ISSN: 2355-9179 the ANOVA analysis with specific comparisons between .
MAJOR vs.
CON, .
TRAIT vs.
CON, and .
MAJOR vs.
TRAIT groups.
Statistical significance was set at a p<0.
All other graphs and plots were generated using Prism 0 (GraphPad Software.
San Diego.
USA).
Results Observed DELs in -thalassaemia A summary of the patients involved in the discovery of lncRNA microarray analysis was listed in Table 1.
Although the discovery sample size was small, significant differences were observed in age and RBC indices (RBC.
Hb.
MCV.
MCH, and hemoglobin analysis (HbA.
HbA2, and HbF) between the groups .
<0.
The summary of differentially expressed lncRNAs (DEL.
with the top five lncRNAs in each comparison was listed in Table 2.
A total of 93 upregulated and 271 downregulated DELs were identified in the MAJOR group compared to the CON group.
In a comparison between the TRAIT and CON groups, there were 10 upregulated and 118 downregulated DELs in the Table 1.
Summary of the clinical information of the individuals used in the microarray Clinical Information Age .
eanASD) CON .
TRAIT .
MAJOR .
p -value 67A5.
67A7.
50A4.
Gender, n (%) Male 1 .
Female 2 .
Malay 3 .
Chinese 1 .
Ethnicity, n (%) RBC .
106/uL) 62A0.
97A0.
90A0.
Hb .
/dL) 63A1.
03A1.
10A0.
Hct (%) 10A2.
13A4.
MCV .
L) 33A4.
00A5.
00A0.
MCH .
20A1.
80A1.
40A0.
MCHC .
/dL) 47 A1.
10A4.
80A0.
RDW-CV (%)
73A0.
30A7.
30A0.
HbA 00A0.
90A0.
40A0.
HbA2 87A0.
20A0.
80A0.
HbF 04A0.
05A0.
60A1.
CE/HPLX (%)
Abbreviation: SD = Standard deviation.
NS = Not significant.
RBC = Red blood cell.
Hb = Hemoglobin.
Hct = Hematocrit.
MCV = Mean corpuscular volume.
MCH = Mean corpuscular hemoglobin.
MCHC = Mean corpuscular hemoglobin concentration.
RDW = Red cell distribution width.
HbA = Hemoglobin A.
HbA2 = Hemoglobin A2.
HbF = Hemoglobin F.
LncRNA Profile in Malaysian -thalassaemia (Mohd Yasin N, et al.
Indones Biomed J.
: 436-46
DOI: 10.
18585/inabj.
Table 2.
List of the top five most dysregulated lncRNAs identified in -thalassaemia major, -thalassaemia trait, and healthy controls comparisons during the discovery lncRNA microarray phase.
Comparison Analysis LncRNAs
Log2
Fold Change LNC-PDCD11-2
LINC01907
MAJOR vs.
CON
93 DELs are upregulated 271 DELs are down-regulated CDRT3
LINC00999
LNC-PTPRU-1
LINC00963
LNC-WWC2-1
FAM74A4
FAM239B
TEX41
LOC101927556
TRAIT vs.
CON
10 DELs are upregulated 118 DELs are downregulated LNC-SEMA6A-5
LNC-C8A-1
LNC-IGIP-1
LNC-TTC26-2
LINC00963
FAM74A4
TEX41
LOC284454
LNC-WWC2-1
LNC-GPR27-1
CDRT3
LINC01907
POT1-AS1
MAJOR vs.
TRAIT
78 DELs are upregulated 22 DELs are downregulated LINC00999
PCBP1-AS1
LNC-DNAJC8-1
CASC15
LNC-PFKP-11
C1orf220 Abbreviation: DELs = Differentially expressed lncRNAs.
MAJOR = -thalassaemia major.
TRAIT = -thalassaemia trait, and CON = healthy controls.
TRAIT group.
In the last comparison of MAJOR vs.
TRAIT
groups, a total of 78 upregulated and 22 downregulated DELs were identified in the MAJOR group compared to the TRAIT group.
Multiple overlapping DELs exist between the comparisons, but none are common for all three comparisons (Figure 1A).
Among these overlapped lncRNAs, 92 lncRNAs were shared between the TRAIT vs.
CON and MAJOR vs.
CON comparisons.
Since these lncRNAs were significant in both thalassaemia groups compared to healthy controls, they may be associated with thalassaemia, regardless of disease severity.
On the other hand, 28 lncRNAs were only present in the TRAIT group compared to the CON group, but not in the other comparisons, which may indicate that these lncRNAs are specifically associated with the trait phenotype.
Between the MAJOR vs.
CON and MAJOR vs.
TRAIT comparisons, we identified 51 common, significantly expressed lncRNAs.
Whereas 40 lncRNAs were only specific to the MAJOR group compared to the TRAIT group, which were not present in other comparisons.
These 91 lncRNAs .
may explain the major phenotype since they are altered in the MAJOR group only, even when compared to the TRAIT group.
Eight lncRNAs were common between TRAIT vs.
CON and MAJOR vs.
TRAIT comparisons.
From these eight lncRNAs, seven of them (LOC101927556.
LNC-SEMA6A-5.
LNC-C8A-1.
LNC-IGIP-1.
EXTL3-AS1.
LNC-ACACA-1, and LNCHMCN1-.
were upregulated in the TRAIT group (Figure 1B) and yet were downregulated in the MAJOR group.
similar pattern was also observed in the last lncRNA.
LNCINPPL1-3, which was reduced in the TRAIT group, yet its expression was higher in the MAJOR group.
These unique differential expression patterns may explain the differences in severity, as these lncRNAs were altered differently in the TRAIT group compared to the MAJOR group.
Significant lncRNAs and Their Predicted miRNA Targets Due to the limited information available on the function of lncRNA and existing databases, a prediction of the target miRNAs that could bind to DELs was performed.
The lncRNA-miRNA relationship is a well-established method to identify the biological functions of these DELs.
From the 92 lncRNAs that are significant in both comparisons between MAJOR vs.
CON and TRAIT vs.
CON, 303
miRNAs were predicted to bind to these 92 lncRNAs.
These 303 miRNAs were then subjected to Gene Ontology (GO) and biological pathway analyses to determine their functions (Figure 2A).
A total of 24 biological processes (BP) were significant, with the top three processes being miRNAmediated post-transcriptional gene silencing, regulatory ncRNA-mediated post-transcriptional gene silencing, and post-transcriptional gene silencing (Supplementary Table For the cellular component (CC) analysis, four CCs were significant: the RISC complex.
RNAi effector complex, ribonucleoprotein complex, and protein-containing No molecular function (MF) was enriched for this set of miRNAs.
The biological pathway analysis The Indonesian Biomedical Journal.
Vol.
No.
October 2025, p.
Print ISSN: 2085-3297.
Online ISSN: 2355-9179 Figure 1.
Summary of the DELs among the three A: The Venn diagram shows the DELs in three comparisons: TRAIT vs.
CON.
MAJOR vs.
CON,
and MAJOR vs.
TRAIT.
B: Heatmap showing the expressions of eight DELs shared between the TRAIT CON and MAJOR vs.
TRAIT comparisons.
values above this threshold are capped values above this threshold are capped -log10.
For the last set of 91 lncRNAs associated with the major phenotype, 245 target miRNAs were predicted and subjected to GO and biological pathways analyses to determine their functions (Figure 2B).
No GO MF was enriched for this set of miRNAs, though 24 BP were significant, with the top three processes being the post-transcriptional gene silencing, miRNA-mediated post-transcriptional gene silencing, and regulatory ncRNA-mediated post-transcriptional gene silencing (Supplementary Table .
For the CC analysis, four CCs were significant: the RISC complex.
RNAi effector complex, ribonucleoprotein complex, and proteincontaining complex.
The biological pathway analysis revealed that only one biological pathway, specifically miRNAs in cancer, was significant according to the KEGG Whereas, five biological pathways were significant based on WikiPathways: the miRNAs involved in DNA damage response, cell differentiation expanded index, glial -log10.
was performed based on three known databases: KEGG.
REACTOME, and WikiPathways.
Only one biological pathway, miRNAs in cancer, was significant based on the KEGG database, whereas two biological pathways were significant based on WikiPathways, which are the miRNAs involved in DNA damage response and 2q21.
1 copy number variation syndrome (Supplementary Table .
For the 28 significant lncRNAs specific to the trait phenotype, 27 miRNAs were predicted as common targets for these lncRNAs.
However, no GO and biological pathways were enriched for these predicted miRNAs.
Similarly, from the eight lncRNAs (Figure 1B) that may explain the severity, seven common miRNAs .
sa-miR-6878-3p, hsa-miR-1205, hsa-miR-10394-5p, hsa-miR-6849-3p, hsa-miR-1224-3p, hsa-miR-5584-5p, and hsa-miR-6873-3.
were predicted to bind to these lncRNAs.
However, no GO enrichment and biological pathways were enriched for these miRNAs.
Figure 2.
Summary of the GO and biological pathway analyses of the predicted microRNA targets of the DELs generated by the go:Profiler tool.
A: GO and biological pathway analyses of the predicted microRNAs from the significant DELs shared between the TRAIT vs.
CON and MAJOR vs.
CON
B: GO and biological pathway analyses of the predicted microRNAs from the significant DELs specific to the MAJOR group.
DOI: 10.
18585/inabj.
cell differentiation, cell differentiation index, and 2q21.
copy number variation syndrome (Supplementary Table .
LncRNA-miRNA Network To further identify the DELs that could explain the disease severity differences between the TRAIT and MAJOR LncRNA Profile in Malaysian -thalassaemia (Mohd Yasin N, et al.
Indones Biomed J.
: 436-46
groups, their DELs and predicted miRNAs were analyzed for their interactions in a molecular network using miRNet and visualized via Cytoscape.
Focusing on the DELs that are altered in the TRAIT group, 45 miRNAs and seven lncRNAs were identified in a molecular network (Figure 3A).
This molecular network was further analyzed for a biological Figure 3.
Molecular networks between the key lncRNAs and their predicted target miRNAs.
The molecular network is specific to lncRNAs and microRNAs associated with Trait phenotypes (A) and Major phenotypes (B).
The blue colour represents lncRNAs, and the orange colour represents predicted miRNAs.
The Indonesian Biomedical Journal.
Vol.
No.
October 2025, p.
pathway analysis to confirm its interactions in disease (Table .
Fifteen biological pathways were enriched for this TRAIT molecular network, with the top three pathways being Cell Proliferation.
Granulopoiesis, and Anti-Cell Proliferation.
Interestingly, one pathway.
Hematopoiesis, was enriched by this network, with the involvement of 15 miRNAs, including the let-7 family, miR-1, miR-15a, and the five key lncRNAs of VASH1-AS1.
PCBP1-AS1.
TYMSOS.
EXTL3-AS1, and LINC00707 (Figure 3A).
Focusing on the DELs altered in the MAJOR group, 76 miRNAs and 25 lncRNAs were constructed in a molecular network (Figure 3B).
A further assessment of this molecular network biological pathway revealed that 11 pathways were significant (Table .
The top three significant pathways were Cell Differentiation.
Anti-Cell Proliferation, and Cell Division.
Interestingly, one pathway.
Glucose metabolism, was only enriched by this network, involving seven miRNAs, including the let-7 family, miR-34a-5p, and three key lncRNAs: PCBP1-AS1.
LINC01001, and VASH1-AS1 (Figure 3B).
Validation of the Significant lncRNAs in -thalassaemia Six potential lncRNAs (VASH1-AS1.
PCBP1-AS1.
EXTL3AS1.
TYMSOS.
LINC01001, and LINC00.
from the molecular networks were further validated in the validation .
In the first comparison (TRAIT vs.
CON),
all the selected lncRNAs were upregulated in the TRAIT individuals .
<0.
(Table .
Only one lncRNA.
TYMSOS, was significant in the MAJOR individuals compared to healthy controls (FC=4.
71, p=0.
A specific comparison between the MAJOR and TRAIT individuals showed that three lncRNAs (VASH1-AS1.
TYMSOS, and LINC01.
were significant, but their expressions were down-regulated in MAJOR individuals compared to TRAIT individuals.
Discussion This study is the first pilot investigation of the long noncoding RNA .
ncRNA) expression profile in Malaysian -thalassaemia major and trait patients.
Previous studies have shown that various lncRNAs are implicated in -thalassaemia patients .
,24,.
, although these studies were performed in different populations.
Despite the limited information available on lncRNAs and their roles in -thalassaemia, the regulatory roles of lncRNAs have been widely reported in other diseases, such as cancer, diabetes, and other blood diseases.
Therefore, there is a need to identify the lncRNAs associated with -thalassaemia.
Print ISSN: 2085-3297.
Online ISSN: 2355-9179 particularly in the population with a high prevalence of -thalassaemia.
In this study, we identified six potential lncRNAs (VASH1-AS1.
PCBP1-AS1.
EXTL3-AS1.
TYMSOS,
LINC01001,
LINC00.
-thalassaemia in Malaysian patients.
Among them, lncRNA TYMSOS was significantly upregulated in the TRAIT individuals, compared to MAJOR and CON individuals.
LncRNA TYMSOS is derived from the opposite strand of the thymidylate synthetase (TYMS) gene that encodes a critical enzyme involved in DNA synthesis and repair.
previous study showed that genetic variants of this TYMS gene contributed to variations in red blood cell folate and homocysteine concentrations in Northwestern European Table 3.
Summary of the significant biological pathways in the molecular networks of trait and major phenotypes.
Biological Pathway Network Hit p -value FDR Cell Proliferation Granulopoiesis Anti-Cell Proliferation Cell Division Hematopoiesis Cell Differentiation Aging Adipocyte Differentiation DNA Damage Response Angiogenesis TRAIT molecular network Cardiotoxicity Cardiac Remodeling Tumor Suppressor MiRNAs T-Cell Differentiation Cell Cycle Cell Differentiation Anti-Cell Proliferation Cell Division T-Cell Differentiation MAJOR molecular network Tumor Suppressor MiRNAs Innate Immunity Aging Adipocyte Differentiation Hematopoiesis Glucose Metabolism Cell Death Abbreviation: MAJOR = -thalassaemia major.
TRAIT = -thalassaemia trait, and FDR = False discovery rate p-value.
LncRNA Profile in Malaysian -thalassaemia (Mohd Yasin N, et al.
Indones Biomed J.
: 436-46
DOI: 10.
18585/inabj.
Table 4.
Summary of the selected differentially expressed lncRNA expressions in the validation cohort.
LncRNA TRAIT vs.
CON
MAJOR vs.
CON
MAJOR vs.
TRAIT
Fold Change p -value Fold Change p -value Fold Change p -value VASH1-AS1
PCBP1-AS1
EXTL3-AS1
TYMSOS
LINC01001
LINC00707
Abbreviation: NS = Not significant.
MAJOR = -thalassaemia major.
TRAIT = -thalassaemia trait, and CON = healthy controls.
Another study on the Chinese population also showed that polymorphism of the TYMS gene was significantly associated with a higher risk of leukemia in individuals with low folate intake .
, suggesting the important role of TYMS in blood regulation and disorders.
Unfortunately, no previous evidence was available on the role of the lncRNA TYMSOS in thalassaemia and blood One transcriptome study of healthy human tissues reported that lncRNA TYMSOS expression was highest in bone marrow tissues, followed by colon and kidney tissues .
, implying a possible role similar to that of its host gene.
TYMS, in blood regulation.
Moreover, in the molecular network analysis, lncRNA TYMSOS was predicted to interact with the miR-214-3p, a known miRNA that regulates oxidative stress in red blood cells.
An in vitro study of oxidative stress in erythroid cells showed that miR-214-3p expression was dose-dependent following hemin treatment .
n oxidative stress induce.
Elevated miR-214-3p promoted oxidative stress in the erythroid cells by suppressing the Activating Transcription Factor 4 (ATF.
, a transcriptional activator of various genes and a key player in cellular stress response.
Based on the known role of miR-214-3p in oxidative stress, the elevated expression of lncRNA TYMSOS observed in the TRAIT group could suggest that lncRNA TYMSOS sponges miR214-3p, acting as a negative feedback mechanism to protect cells from oxidative stress.
Furthermore, this negative regulation of lncRNA TYMSOS towards miR-214-3p may also partly explain the major phenotype, as the expression of this lncRNA TYMSOS was reduced in the MAJOR group, implying that this negative regulation may be absent or However, further studies are needed to confirm this relationship between lncRNA TYMSOS and miR-2143p, particularly their interactions that could explain the differences between major and trait phenotypes.
Besides lncRNA TYMSOS, two lncRNAs (VASH1AS1 and LINC01.
were altered in the MAJOR group compared to the TRAIT group.
Similar to TYMSOS, both lncRNAs have no previous publication of their roles in blood disorders, particularly thalassaemia.
However, in the molecular network analysis, both lncRNAs were shown to co-regulate the let-7 miRNA family.
A previous study of pediatric thalassaemia in the Chinese population showed that let-7 miRNAs, hsa-let-7f-1-3p and hsa-let-7a-3p, were upregulated, whereas hsa-let-7b-5p, hsa-let-7i-5p, and hsa-let-7g-5p were downregulated in patients with thalassaemia compared to healthy controls.
Moreover, the same study also showed that two of these miRNAs, hsa-let-7b-5p and hsa-let-7i-5p, were significantly reduced in adolescent and adult patients.
Importantly, these let7 miRNAs were shown to regulate the B-cell lymphoma/ leukemia 11A (BCL11A) gene, a critical regulator of foetal-to-adult hemoglobin switching, and have been implicated in -thalassaemia.
Since a high level of fetal hemoglobin (HbF) .
p to 90%) is often observed in -thalassaemia major compared to -thalassaemia trait, the co-regulation of let-7 miRNAs by VASH1-AS1 and LINC01001 lncRNAs may partly suggest that they are involved in hemoglobin switching, thus contributing to the differences seen in the two phenotypes.
Therefore, further studies are needed to explore the relationship between VASH1-AS1 and LINC01001 lncRNAs with the let-7 family, particularly in contributing to the dysregulation of foetal-toadult hemoglobin switching.
Unfortunately, the sample size for the discovery and validation was small, and there was an imbalance between the groups.
These issues may affect the statistical power to identify truly significant lncRNAs and increase the risk of Type II errors, thus potentially impacting the reproducibility of the results.
Furthermore, some of these lncRNAs exhibit The Indonesian Biomedical Journal.
Vol.
No.
October 2025, p.
opposite expression between the major and trait individuals, which may be due to actual biological changes in disease however, these changes could also be influenced by the small sample size.
Moreover, there are potential confounding factors that could contribute to changes in lncRNA expression, such as age, gender, ethnicity, presence of other co-morbidities, and environmental factors, which may affect the differences observed in the findings.
Previous studies of Malaysian -thalassaemia patients reported that factors such as age and BTB and CNC homology 1 (BACH.
expression were associated with clinical phenotypes .
however, whether these factors affect lncRNA expression requires further investigation with a larger cohort with a much more diverse population.
Another limitation is that all the regulatory networks of lncRNAs and miRNAs are based on prediction analysis and have not been confirmed through direct binding experiments, such as luciferase assays.
We also did not measure the expressions of target miRNAs in the validation phase, thus limiting the interpretations of these molecular networks and their contributions to the differences seen in MAJOR and TRAIT Therefore, future studies should incorporate the validation of miRNA expression in the same cohort and confirm their direct molecular binding.
Conclusion Three key lncRNAs (TYMSOS.
VASH1-AS1, and LINC01.
were significantly altered between MAJOR and TRAIT individuals, which may partly explain the phenotypes seen in the Malaysian -thalassemia patients.
However, the current findings would require further validation with a larger and more diverse cohort to confirm the significance of these lncRNAs, as well as functional studies to evaluate the direct interactions and regulation of these molecules.
Nevertheless, the identification of these lncRNAs may serve as new biomarkers to distinguish disease severity in -thalassaemia patients and contribute to the future development of a diagnostic and prognostic panel for these patients.
Acknowledgments We thank the Director General of Health Malaysia for his permission to publish this article.
This manuscript is funded by the Institute for Medical Research (IMR) of Malaysia (Grant code: NMRR-19-2280-48.
The funders have Print ISSN: 2085-3297.
Online ISSN: 2355-9179 no role in the drafting, writing, and publishing of the Authors Contribution NMY and SAS proposed the study design and concept, performed data analyses, results interpretation and wrote the manuscript.
NMY secured the grant.
NFAR and NK performed sample collection, microarray, and Real-Time PCR experiments, as well as analysed the data.
NFAR
wrote the first draft of the manuscript, while NAAM.
ENS.
YY.
HMI.
ZM.
EE reviewed and edited the final draft.
All authors took part in a critical revision of the manuscript.
Conflict of Interest All authors declare that they have no competing interests.
References