Changes in lncRNAs and related genes in β-thalassemia minor and β-thalassemia major

Abstract

β-thalassemia is caused by β-globin gene mutations. However, heterogeneous phenotypes were found in individuals with same genotype, and still undescribed mechanism underlies such variation. We collected blood samples from 30 β-thalassemia major, 30 β-thalassemia minor patients, and 30 matched normal controls. Human lncRNA Array v2.0 (8 × 60 K, Arraystar) was used to detect changes in long non-coding RNAs (lncRNAs) and mRNAs in three samples each from β-thalassemia major, β-thalassemia minor, and control groups. Compared with normal controls, 1424 and 2045 lncRNAs were up- and downregulated, respectively, in β-thalassemia major patients, whereas 623 and 349 lncRNAs were up- and downregulated, respectively, in β-thalassemia minor patients. Compared with β-thalassemia minor group, 1367 and 2356 lncRNAs were up- and downregulated, respectively, in β-thalassemia major group. We selected five lncRNAs that displayed altered expressions (DQ583499, X-inactive specific transcript (Xist), lincRNA-TPM1, MRFS16P, and lincRNA-RUNX2-2) and confirmed their expression levels in all samples using real-time polymerase chain reaction. Based on coding-non-coding gene co-expression network and gene ontology biological process analyses, several signaling pathways were associated with three common organ systems exhibiting β-thalassemia phenotypes: hematologic, skeletal, and hepatic systems. This study implicates that abnormal expression levels of lncRNAs and mRNA in β-thalassemia cases may be correlated with its various clinical phenotypes.

Keywords

β-thalassemia / long non-coding RNA / mRNA / phenotypic heterogeneity / pathway

Introduction

b-thalassemia is common monogenic life-threatening blood disorder and poses persistent threat to public health. Approximately 4.5% of global population carries b-thalassemia mutation [ 1]. Particularly high prevalence of mutation occurs in populations in Mediterranean, the Middle East, Transcaucasia, Indian subcontinent, and the Far East; b-thalassemia is relatively common in populations of African descent, in China, Japan, Northern Europe, North and South Americas, the Caribbean, and Australia [ 2].

b-thalassemia belongs to heterogeneous group of hemoglobin synthesis disorders, wherein mutations reduce synthesis or stability of b-globin chain. Three clinical and hematological types are recognized: b-thalassemia minor or carrier state (Tr) results from heterozygous mutation of b-globin gene and is clinically asymptomatic or manifests as mild anemia and defined by specific hematological characteristics; b-thalassemia major (Tm) is severe transfusion-dependent anemia caused by homozygous or compound heterozygous mutation of b-globin gene and has several clinical manifestations, including facies, jaundice, hepatosplenomegaly, growth retardation, osteoporosis, and pathological fracture; b-thalassemia intermedia (Ti) has clinical severity that lies between Tr and Tm [ 2].

Although b-thalassemia is caused by b-globin gene mutations, complete one-to-one correspondences is not observed between these gene mutations and their clinical phenotypes. Tm and Ti comprise clinically and genotypically heterogeneous groups of thalassemia-like disorders, ranging in severity, from mild to severe transfusion-dependent states [ 3]. In addition to phenotype varieties that result from allelic heterogeneity at b-globin locus, similar b-globin gene mutation induces constantly changing phenotype, which cannot be explained by mutation profile [ 4, 5]. This clinical heterogeneity leads to difficulty in prevention, prenatal screening, clinical diagnosis, and treatment of b-thalassemia.

Available information can explain in detail molecular mechanisms underlying b-thalassemia heterogeneity. Recently, modifier genes were discovered to adjust hemoglobin peptide synthesis and balance, thereby influencing clinical phenotype to varying extent [ 6]. In addition to gene modification, non-coding RNAs are implicated in progression of globin gene expression, particularly in g-globin gene expression reactivation in association with increased fetal hemoglobin (HbF) synthesis [ 7]. MicroRNAs (miRNAs) may contribute to HbF level variation in whole blood of b-thalassemia patients and clinical severity of b-thalassemia [ 8]. In recent years, long non-coding RNA (lncRNA) became popular topic among epigenetics researches. lncRNAs are RNA species>200 bp in length, and they are frequently polyadenylated, devoid of evident open reading frames, and are transcribed by RNA polymerase II [ 9]. Thousands of lncRNAs in mammalian genome play functional roles in cells [ 10]. lncRNA-mediated activity is implicated in various cellular processes and is important in both human development and diseases [ 11]. Roles are emerging for lncRNAs in ribonucleic protein complexes that regulate various stages of gene expression. Intrinsic nucleic acid nature of lncRNAs confers their dual ability to function as ligands for proteins (such as those that function in gene regulation) and to mediate base-pairing interactions that guide lncRNA-containing complexes to specific RNA or DNA target sites. However, research does not explore mechanisms by which lncRNAs identify their genomic targets [ 12]. Dysregulation of lncRNAs was associated with many human diseases, including various kinds of cancers [ 13], such as hepatocellular carcinoma [ 14], renal clear cell carcinoma [ 15], and ovarian cancer [ 16]. In cancer tissue, lncRNAs often become prominent components of transcriptional regulation by associating with epigenetic complexes [ 17, 18]. However, although past studies showed that some lncRNAs participate in regulatory circuitry underlying erythropoiesis [ 19], which is related to b-thalassemia, little is known about how lncRNAs control b-globin gene transcription. We speculated that the change of lncRNAs may regulate b-globin gene expression, specifically affecting b-globin synthesis. Moreover, mutated b-globin gene may affect expression of associated lncRNAs and ultimately change target gene expressions.

We determined expression profiles of lncRNAs in samples from Tr and Tm patients and normal controls (NCs) and observed differential expression of numerous lncRNAs. Our results demonstrated that lncRNA expression profile may provide new molecular biomarkers for diagnosing b-thalassemia. Influence of some lncRNA on b-globin gene regulation may constitute novel therapeutic target to improve curative effects on b-thalassemia patients.

Materials and methods

Patient samples

Written informed consent was obtained from all patients and NC subjects. This study was approved by Institutional Review Board of Bao’an Maternal and Children Health Hospital in Shenzhen. This study included 30 Tm patients undergoing regular blood transfusion treatment, 30 Tr patients undergoing pre-pregnancy evaluation, and 30 NCs who were present in physical examinations. Microarray analyses of lncRNAs and mRNAs utilized samples from three Tm patients, three Tr patients, and three matching control subjects, and all samples were used for further evaluation. b-thalassemia was diagnosed per genetic assessment combined with hematological indices and clinical characteristics. Whole blood samples from each subject were snap-frozen in a -80 °C freezer after collection. Table S1 summarizes detailed information of three Tm and three Tr patients included in the microarray set.

RNA extraction

Total RNA was extracted from peripheral blood using TRIzol reagent (Invitrogen, Carlsbad, California, USA) per manufacturer’s protocol. Total RNA from each sample was quantified using NanoDrop ND-2000 spectrophotometer. RNA integrity was assessed via standard denaturing agarose gel electrophoresis.

Microarray and computational analyses

For microarray analysis, Agilent Array platform was employed. Sample preparation and microarray hybridization were performed based on manufacturer’s standard protocols with minor modifications. mRNA was purified from total RNA after removal of rRNA (mRNA-ONLY™ Eukaryotic mRNA Isolation Kit, Epicenter). Each sample was amplified and transcribed into fluorescent cRNA along entire length of transcripts without 3′ bias using random priming method. Labeled cRNAs were hybridized to Human lncRNA Array v2.0 (8 × 60K, Arraystar). After washing the slides, arrays were scanned using Agilent G2505C Scanner.

Arraystar Human lncRNA Microarray v2.0 is designed for global profiling of human lncRNAs and protein-coding transcripts. Second-generation lncRNA microarray detected 33 045 lncRNAs and 30 215 coding transcripts. These lncRNAs were carefully collected from authoritative databases, such as RefSeq, UCSC Knowngenes, and Ensembl, and many related articles. Each transcript is represented by specific exon or splice junction probe that can accurately identify individual transcripts. Positive probes for housekeeping genes and negative probes are printed onto array for hybridization quality control.

Agilent Feature Extraction software (version 11.0.1.1) was used to analyze acquired array images. Quantile normalization and subsequent data processing were performed using GeneSpring GX v11.5.1 software package (Agilent Technologies). After quantile normalization of raw data, further data analysis involved lncRNAs and mRNAs, which were detected in at least three out of nine samples and annotated as present or marginal (“All Targets Value”). Significantly differentially expressed lncRNAs and mRNAs were identified via Volcano Plot filtering. Pathway and gene ontology (GO) analyses were applied to determine roles of these differentially expressed mRNAs in biological pathways or GO terms. Finally, hierarchical clustering was performed to display distinct lncRNA and mRNA expression patterns between samples.

Construction of coding-non-coding gene co-expression (CNC) network

Network construction procedures included the following: (1) data preprocessing: for same coding gene, median value of different transcripts represents gene expression value without special treatment of lncRNA expression values; (2) data screening: assessing differential expressions of lncRNA and mRNA; (3) calculation and removal of subset of data according to Pearson correlation coefficient (PCC) and calculation of correlation coefficient of PCC between lncRNA coding genes using R values; (4) screening based on PCC, selecting sequences displaying PCC<0.99 as meaningful subset, and constructing CNC network using Cytoscape. V indicates each lncRNA, and circular nodes represent mRNAs. lncRNAs are labeled in eight colors per number of mRNAs connected to lncRNA. Node size represents gene expression: larger dot denotes higher expression value. Lines represent gene co-expression relationship between specific lncRNA and mRNA; dark gray lines represent positive correlation, and light gray lines indicate negative correlation.

Quantitative polymerase chain reaction (qPCR) of five upregulated or downregulated lncRNAs

Total RNA was extracted from fresh peripheral blood using TRIzol reagent (Invitrogen, Carlsbad, California, USA) and reverse transcribed using Takara Primescript RT Reagent Kit with gDNA Eraser (Perfect Real Time) per manufacturer’s instructions. lncRNA expression in blood was measured via qPCR using SYBR Premix Ex Taq and ViiA 7 Real-time PCR System (Applied Biosystems, Foster, California, USA). This study used the following primers: glyceraldehyde 3-phosphate dehydrogenase (GAPDH): F: 5′GGGAAACTGTGGCGTGAT3′; R: 5′GAGTGGGTGTCGCTGTTGA3′; X-inactive specific tr-anscript (Xist): F: 5′GGCTTAGGGCTAGTCGTTTGT3′; R: 5′TTCCTCTGCCTGACCTGCTAT3′; lincRNA-TPM1: F: 5′CACATCTTAGCCACGGAACACT3′; R: 5′AAAGCAGGACAGACATCAGCAC3′; lincRNA-RUNX2: F: 5′TAGTCCTTGCTGTCCACCACC3′; R: 5′ CCTGGAATGCTGACTGGGTAT3′; DQ583499: F: 5′GGCTACATGACAGAGGGTGCA3′; R: 5′TCCAGGCCAGGAGTTGAGAA3′; MRPS16P: F: 5′AACCAGCCCTTCTACCGCA3′; R: 5′TCGAACC-AGCAAGACCCACT3′. Five lncRNAs (DQ583499, Xist, lincRNA-TPM1, MRFS16P, and lincRNA-RUNX2-2) were selected from significantly differentially expressed lncRNAs, which were evaluated in 90 samples. For each sample, 1 mg total RNA was reverse transcribed into cDNA per manufacturer’s protocol. qPCR was performed in total reaction volume of 20 µl, consisting of 10 µl SYBR Premix Ex Taq (Tli RNaseH Plus) (2×), 0.8 µl forward PCR primer (10 μmol/L), 0.8 µl reverse PCR primer (10 μmol/L), 0.4 µl ROX Reference Dye II (50×), 2 µl cDNA, and 6 µl RNA-free water. Real-time qPCR reaction parameters included initial denaturation step of 10 min at 95 °C followed by 40 cycles at 95 °C (10 s), 60 °C (60 s), and 72 °C (30 s), followed by final extension at 72 °C for 5 min. All experiments were conducted in triplicate. All samples were normalized to GAPDH. Median of each triplicate was used to calculate relative lncRNA concentrations (DCt= Ct median lncRNA − Ct median GAPDH).

Statistical analysis

Fold-changes in expression were calculated using 2^-ΔΔCt method. Expression differences in lncRNAs between b-thalassemia and control samples were analyzed using SPSS 17.0 (SPSS, Inc., Chicago, IL, USA). P<0.05 was considered significant. Statistical analyses included one-way analysis of variance, Fisher’s exact test, and two-tailed Student’s t-test.

Results

Overview of lncRNA profiles

Based on lncRNAs expression profiles of three Tr, three Tm, and three NC samples, specific lncRNAs were differentially expressed (Fig. 1). In paired samples, lncRNA expression profiles were compared by calculating log fold-change in expression profiles (Tm/Tr/NC). Comparison was performed as follows: fold-change threshold: 2.0; a positive value indicates upregulation and negative value indicates downregulation. Log fold-change represents log₂ value of absolute fold-change. Fold-changes and P values were calculated from normalized expression levels. We identified thousands of differentially expressed human lncRNAs in RefSeq_NR, UCSC_knowngene, Ensembl, H-invDB, Fantom, Fantom_stringent, NRED, RNAdb, misc_lncRNA, UCR, and lncRNA databases.

Next, based on microarray data, we compared lncRNA expression levels among three Tm patients, three Tr patients, and three matching NCs. Results demonstrated that tens of thousands of lncRNAs could be detected in normal and β-thalassemia blood, but only thousands of them were significantly upregulated or downregulated; these distinct expression profiles distinguished β-thalassemia patient blood from matching NC blood. Compared with NC samples, 1424 and 2045 lncRNAs were upregulated and downregulated, respectively, in all three Tm samples; 623 and 349 lncRNAs were found to be upregulated and downregulated, respectively, in all three Tr samples; 1367 and 2356 lncRNAs were upregulated and downregulated, respectively, in all three Tm samples compared with three Tr samples. Most of these lncRNAs are not functionally characterized. Between Tm patients and NCs, Xist (Log₂ fold-change (Tm/NC) = 378.6) was the most significantly upregulated lncRNA, whereas BC039532 (Log₂ fold-change (Tm/NC) = 35.6) was the most significantly downregulated. Between Tr patients and NCs, NR_002312 (Log₂ fold-change (Tr/NC) = 57.9) was the most significantly upregulated lncRNA, whereas NR_015392 (Log₂ fold-change (Tr/NC) = 44.9) was the most significantly downregulated. Upregulated and downregulated lncRNAs varied in number between different groups (Tables 1–3). Downregulated lncRNAs were more commonly detected than upregulated lncRNAs.

Overview of mRNA profiles

Up to 13 664 coding transcripts were detected in three groups using 30 215 coding transcript probes. In three Tm samples, averages of 2408 and 1407 mRNAs were upregulated and downregulated, respectively, compared with matching NC blood samples. In three Tr samples, averages of 637 and 146 mRNAs were upregulated and downregulated, respectively, compared with matching NC blood samples. In three Tm samples, averages of 1734 and 890 mRNAs were upregulated and downregulated, respectively, compared with Tr samples (Tables 4‒6). GO and pathway analyses revealed that differentially expressed mRNAs may be involved in cell differentiation, cellular development, chemical stimulus response, system development, multicellular organismal process, and metabolic process-associated signal pathway. These results support the notion that b-thalassemia is metabolic and immune disease.

Real-time qPCR validation

Future studies must focus on larger survey samples at individual level. We examined expression levels of five lncRNAs (DQ583499, MRFS16P, Xist, lincRNA-TPM1, and lincRNA-RUNX2-2) in 30 Tm, 30 Tr, and 30 NC samples, including chip test samples using qPCR. Results were consistent with previous microarray results (Figs. 2 and 3) and demonstrated that Xist, lincRNA-TPM1, and lincRNA-RUNX2-2 were upregulated, and DQ583499 and MRFS16P were downregulated in b-thalassemia samples compared with NC samples (P <0.001 for each lncRNAs).

Construction of CNC network

CNC network was constructed based on correlation analysis between differentially expressed lncRNAs and mRNAs. mRNAs and lncRNAs displaying PCCs equal to or greater than 0.99 were selected to construct network using Cytoscape program. Within this co-expression network, nine lncRNAs and 273 mRNAs composed CNC network node. Most pairs of co-expressed lncRNAs and mRNAs displayed positive correlation. CNC network indicated that one lncRNA regulated multiple mRNAs, and one mRNA was regulated by multiple lncRNAs (Fig. 4). CNC network underscored that lncRNAs and mRNAs may have regulatory relationships and may play important role in b-thalassemia pathogenesis.

GO enrichment analysis

GO pathway analysis revealed that differentially expressed mRNAs may be associated with specific biological processes, including response to hypoxia, muscle organ development, signal transduction, negative caspase activity regulation, apoptosis induction by extracellular signals, mitogen-activated protein kinase (MAPK) activation, chemotaxis, apoptosis induction, cytokine-mediated signaling pathway, inflammatory response, synaptic transmission, and anti-apoptosis pathways (Fig. 5). Moreover, further genetic analysis demonstrated that based on CNC network, our selected genes were upregulated or downregulated in single or multiple pathways and involved in development of hematologic, skeletal, and hepatic systems, which are major internal organ systems affected by b-thalassemia (Tables 7 and 8). These results suggest that b-thalassemia has various genetic and phenotypic characteristics.

Discussion

b-thalassemia is among the most prevalent hereditary diseases and presents public health problem worldwide. Although subgroup of b-thalassemia alleles presents dominant negative mutations, most b-thalassemia alleles are inherited in Mendelian recessive manner [ 20]. b-thalassemia is caused by mutations in b-globin chain of hemoglobin and characterized by quantitative deficiency of b-globin chains, causing ineffective red blood cell production and anemia [ 20]. Homozygotes and compound heterozygotes of b-thalassemia allele exhibit varied clinical manifestations from severe transfusion-dependent disorder from early infancy or Tm to relatively milder Ti phenotype. This diversity is primarily related to globin chain synthesis imbalance, which can be due to nature of b-globin gene mutations or interaction between mutant gene and a or g-loci [ 21]. b-thalassemia phenotypes cannot be solely explained by b-globin gene mutation because similar mutation occasionally produces significantly distinct phenotype. Epigenetic mechanisms, including DNA methylation and histone acetylation, may be involved in pathogenesis of Mediterranean anemia [ 22]. Epigenetic phenomena tend to be stochastic, reversible, and mosaic; their rules of occurrence and inheritance of epimutations are different from Mendelian genetics [ 23]. In addition to b-globin gene mutations, hemoglobin peptide synthesis or balance is influenced by other modifying genes, such as BCL11A [ 24], GATA1 [ 25], KLF1 [ 26], and HDAC1 [ 27].

No previous report described expression profiles of lncRNAs in thalassemia, nor examined association between lncRNA expression and clinical characteristics or outcomes of b-thalassemia. We analyzed three Tr, three Tm, and three NC blood samples using microarray techniques and selected five lncRNAs for validation via qPCR in 90 samples. Using abundant and varied probes accounting for 33 045 lncRNAs and 30 215 coding transcripts in microarray, we quantitatively determined expression levels of thousands of lncRNAs. In Tr and Tm samples, expression profiles of these lncRNAs were thoroughly analyzed to investigate role of lncRNAs in development of different b-thalassemia types. Based on microarray data, 28 443 lncRNAs were expressed, and thousands of lncRNAs were upregulated or downregulated when compared pairwise with three groups. Most of these lncRNAs are not functionally characterized. Between Tm and NC samples, lncRNAs presented remarkable differential expressions, whereas those between Tr and NC samples were less clear. These findings may provide potential strategy to distinguish between b-thalassemia and normal blood.

Most lncRNAs display spatial and temporal specificities and are expressed during organismal differentiation and development. Characteristic profiles of lncRNA expression were described in prostate carcinoma, hepatic tumors, and placentas of preeclampsia fetuses. These lncRNAs may be exerted during specific cellular functions in development of b-thalassemia and may provide novel insight into molecular basis of this disease.

According to changed folds of expression levels and correlation among clinical phenotypes of b-thalassemia, five differentially expressed lncRNAs (DQ583499, MRFS16P, Xist, lincRNA-TPM1, and lincRNA-RUNX2-2) were selected for further validation via qPCR of 30 samples for each group. Analysis results coincided with those of microarray assay.

Many lncRNAs regulate chromatin, but mechanisms by lncRNA localization to genomic targets remain unexplored. lncRNA DQ583499 is 401 bp intergenic lncRNA transcribed from gene ENSG00000214145 (RP11-513G11.1), which is located on chromosome 19q. lncRNA Xist is 32 094 bp intergenic lncRNA on chromosome Xq13.2. Xist is expressed in mammalian females and transcriptionally silences one of the pair of X during early development, thereby providing dosage equivalence of genes on X chromosome between males and females. Xist transcript is spliced RNA. Alternatively spliced transcriptional variants were identified, but their full-length sequences were not determined. During maintenance of X-chromosome inactivation (XCI), Xist binds broadly across X chromosome. During XCI initiation, Xist initially binds to distal regions across X chromosomes that are undefined by specific sequences. Xist identifies these regions by exploiting three-dimensional conformations of X chromosome and requires silencing domain to spread across actively transcribed regions to access entire chromosome [ 28]. lncRNA MRFS16P is 294 bp intergenic lncRNA, which is transcribed from MRFS16P gene on chromosome 20q13.32. This gene is MRFS16 pseudogene. MRFS16 is similar to MRF3 and MRF4 genes and encodes for nuclear protein belonging to myogenic factor subfamily of basic helix-loop-helix family of transcription factors. MRFS16P regulates muscle cell differentiation by cell cycle arrest, which is prerequisite for myogenic initiation. Moreover, MRFS16P is involved in muscle regeneration. This gene activates its own transcription and possibly stabilizes commitment to myogenesis. MRF4 and MRF3 play competitive roles in myogenesis by stimulating cell proliferation and differentiation [ 29]. Furthermore, MRFs are differentially required for induction of miRNA gene expression during skeletal myogenesis [ 30]. lncRNA-TPM1 is 9049 bp intergenic lncRNA transcribed from TPM1 gene on chromosome 15. TPM1 widely distributes actin-binding proteins involved in contractile system of striated and smooth muscle cells and cytoskeleton of non-muscle cells [ 31]. Mutations in this gene are associated with type 3 familial hypertrophic cardiomyopathy [ 32]. TPM1 gene is member of highly conserved tropomyosin (TPM) family of class II tumor suppressor genes. Sequences of TPM family genes are structurally intact but are under- or non-expressed because of downregulation or silencing in transcription or translation [ 33]. TPMs are widely distributed in all cell types associated with actin, and they serve as actin-binding proteins that stabilize microfilaments. Suppression of TPM1 and TPM2 expression was reported in malignant cells, suggesting role of these proteins in neoplastic transformation. TPMs also regulate both microfilament organization and anchorage-independent growth, emphasizing importance of TPMs in cell transformation [ 34]. lncRNA-RUNX2-2 is 361 bp intergenic lncRNA transcribed from RUNX2 gene on chromosome 6p21. RUNX2 gene is member of RUNX family of transcription factors and encodes nuclear protein containing Runt-DNA-binding domain. This protein is essential for osteoblastic differentiation and skeletal morphogenesis and acts as scaffold for nucleic acids and regulatory factors involved in skeletal gene expression. RUNX2 protein acts as subunit of heterodimeric complex, wherein both monomers bind to DNA at high affinity. Mutations in RUNX2 are associated with bone development disorder cleidocranial dysplasia [ 35].

To gain insight into functions of targets of lncRNAs, GO and KEGG enrichment analyses were applied to selected genes based on correlation analyses among nine and 273 differentially expressed lncRNAs and mRNAs, respectively. Based on this analysis, we identified specific pathways, such as response to hypoxia and muscle organ development pathway, that are related to clinical phenotype of b-thalassemia, indicating involvement of lncRNAs and mRNAs. These lncRNAs and mRNAs may play roles in emergence of histanoxia, which is caused by elevated HbF levels and dysontogenesis in Tm patients. Furthermore, after assessing genes and gene products associated with these pathways, we noted that expression levels of involved genes were either upregulated or downregulated in single or multiple pathways, indicating that lncRNAs may act regulating protein-coding genes in such pathways. Moreover, because clinical phenotype of b-thalassemia primarily manifests in hematologic and skeletal systems and liver and spleen, we analyzed genes regulating development of organs mentioned above. Results revealed that some of our target genes participated in processes related to hematologic, skeletal, and hepatic system developments, thereby suggesting role of lncRNAs in b-thalassemia pathogenesis. Among these genes, for example, EGFR was shown to be expressed in liver macrophages of hepatocellular carcinoma patients with poor survival [ 36]. Moreover, aberrant regulation of FGF signaling pathways, including FGF8, may lead to various skeletal anomalies [ 37]. HSPA1B (70 kDa) is more upregulated in myelodysplastic syndrome patients compared with controls [ 38]; increased SOD1 activity was found in patients with hematological malignancies [ 39]. However, we did not find any target genes related to spleen development, which may be due to specific set of lncRNAs that we selected for analysis. Additional lncRNAs will be selected for examination in further studies.

To the best of our knowledge, this study was the first to describe expression profiles of human lncRNAs in b-thalassemia using microarray analysis. Many lncRNAs were aberrantly downregulated in Tm samples compared with Tr and matched NC samples. These downregulated lncRNAs possibly play key roles as disease loci during b-thalassemia development and progression. Additional research is needed to determine whether these lncRNAs serve as novel therapeutic targets or diagnostic biomarkers in b-thalassemia patients.

References

[1] Tuzmen S, Schechter AN. Genetic diseases of hemoglobin: diagnostic methods for elucidating β-thalassemia mutations. Blood Rev 2001; 15(1): 19–29

[2] Cao A, Galanello R. β-thalassemia. Genet Med 2010; 12(2): 61–76

[3] Galanello R, Ruggeri R, Paglietti E, Addis M, Melis MA, Cao A. A family with segregating triplicated α globin loci and β thalassemia. Blood 1983; 62(5): 1035–1040

[4] Harteveld CL, Refaldi C, Cassinerio E, Cappellini MD, Giordano PC. Segmental duplications involving the α-globin gene cluster are causing β-thalassemia intermedia phenotypes in β-thalassemia heterozygous patients. Blood Cells Mol Dis 2008; 40(3): 312–316

[5] Sollaino MC, Paglietti ME, Perseu L, Giagu N, Loi D, Galanello R. Association of α globin gene quadruplication and heterozygous β thalassemia in patients with thalassemia intermedia. Haematologica 2009; 94(10): 1445–1448

[6] Viprakasit V, Tanphaichitr VS, Chinchang W, Sangkla P, Weiss MJ, Higgs DR. Evaluation of α hemoglobin stabilizing protein (AHSP) as a genetic modifier in patients with β thalassemia. Blood 2004; 103(9): 3296–3299

[7] Sankaran VGMT, Menne TF, Šćepanović D, Vergilio JA, Ji P, Kim J, Thiru P, Orkin SH, Lander ES, Lodish HF. MicroRNA-15a and-16-1 act via MYB to elevate fetal hemoglobin expression in human trisomy 13. Proc Natl Acad Sci USA 2011; 108(4): 1519–1524

[8] Lee YT, de Vasconcellos JF, Yuan J, Byrnes C, Noh SJ, Meier ER,  Kim KS,  Rabel A,  Kaushal M,  Miller JL. LIN28B-mediated expression of fetal hemoglobin and production of fetal-like erythrocytes from adult human erythroblasts ex vivo. Blood 2013;122(6):1034–1041

[9]  .Prensner JR, Chinnaiyan AM. The emergence of lncRNAs in cancer biology. Cancer Discov 2011; 1(5): 391–407

[10] Guttman M, Donaghey J, Carey BW, Garber M, Grenier JK, Munson G, Young G, Lucas AB, Ach R, Bruhn L, Yang X, Amit I, Meissner A, Regev A, Rinn JL, Root DE, Lander ES. lincRNAs act in the circuitry controlling pluripotency and differentiation. Nature 2011; 477(7364): 295–300

[11] Ponting CP, Oliver PL, Reik W. Evolution and functions of long noncoding RNAs. Cell 2009; 136(4): 629–641

[12] Guttman M, Rinn JL. Modular regulatory principles of large non-coding RNAs. Nature 2012; 482(7385): 339–346

[13] Wapinski O, Chang HY. Long noncoding RNAs and human disease. Trends Cell Biol 2011; 21(6): 354–361

[14] Yang F, Zhang L, Huo XS, Yuan JH, Xu D, Yuan SX, Zhu N, Zhou WP, Yang GS, Wang YZ, Shang JL, Gao CF, Zhang FR, Wang F, Sun SH. Long noncoding RNA high expression in hepatocellular carcinoma facilitates tumor growth through enhancer of zeste homolog 2 in humans. Hepatology 2011; 54(5): 1679–1689

[15] Yu G, Yao W, Wang J, Ma X, Xiao W, Li H, Xia D, Yang Y, Deng K, Xiao H, Wang B, Guo X, Guan W, Hu Z, Bai Y, Xu H, Liu J, Zhang X, Ye Z. lncRNAs expression signatures of renal clear cell carcinoma revealed by microarray. PLoS ONE 2012; 7(8): e42377

[16] Akrami R, Jacobsen A, Hoell J, Schultz N, Sander C, Larsson E. Comprehensive analysis of long non-coding RNAs in ovarian cancer reveals global patterns and targeted DNA amplification. PLoS One 2013; 8(11): e80306

[17] Tsai MC, Manor O, Wan Y, Mosammaparast N, Wang JK, Lan F, Shi Y, Segal E, Chang HY. Long noncoding RNA as modular scaffold of histone modification complexes. Science 2010; 329(5992): 689–693

[18] Rinn JL, Kertesz M, Wang JK, Squazzo SL, Xu X, Brugmann SA, Goodnough LH, Helms JA, Farnham PJ, Segal E, Chang HY. Functional demarcation of active and silent chromatin domains in human HOX loci by noncoding RNAs. Cell 2007; 129(7): 1311–1323

[19] Alvarez-Dominguez JR, Hu W, Yuan B, Shi J, Park SS, Gromatzky AA, van Oudenaarden A, Lodish HF. Global discovery of erythroid long noncoding RNAs reveals novel regulators of red cell maturation. Blood 2014; 123(4): 570–581

[20] Thein SL. The molecular basis of β-thalassemia. Cold Spring Harb Perspect Med 2013; 3(5): a011700

[21] Colah R, Nadkarni A, Gorakshakar A, Phanasgaonkar S, Surve R, Subramaniam PG, Bondge N, Pujari K, Ghosh K, Mohanty D. Impact of β globin gene mutations on the clinical phenotype of beta thalassemia in India. Blood Cells Mol Dis 2004; 33(2): 153–157

[22] Kim A, Kiefer CM, Dean A. Distinctive signatures of histone methylation in transcribed coding and noncoding human β-globin sequences. Mol Cell Biol 2007; 27(4): 1271–1279

[23] Martin DI, Ward R, Suter CM. Germline epimutation: a basis for epigenetic disease in humans. Ann N Y Acad Sci 2005; 1054(1): 68–77

[24] Sankaran VG, Menne TF, Xu J, Akie TE, Lettre G, Van Handel B, Mikkola HK, Hirschhorn JN, Cantor AB, Orkin SH. Human fetal hemoglobin expression is regulated by the developmental stage-specific repressor BCL11A. Science 2008; 322(5909): 1839–1842

[25] Jawaid K, Wahlberg K, Thein SL, Best S. Binding patterns of BCL11A in the globin and GATA1 loci and characterization of the BCL11A fetal hemoglobin locus. Blood Cells Mol Dis 2010; 45(2): 140–146

[26] Zhou D, Liu K, Sun CW, Pawlik KM, Townes TM. KLF1 regulates BCL11A expression and γ- to β-globin gene switching. Nat Genet 2010; 42(9): 742–744

[27] Thein SL. Genetic association studies in β-hemoglobinopathies. Hematology Am Soc Hematol Educ Program 2013;354–361

[28] Engreitz JM, Pandya-Jones A, McDonel P, Shishkin A, Sirokman K, Surka C, Kadri S, Xing J, Goren A, Lander ES, Plath K, Guttman M. The Xist lncRNA exploits three-dimensional genome architecture to spread across the X chromosome. Science 2013; 341(6147): 1237973

[29] Jin X, Kim JG, Oh MJ, Oh HY, Sohn YW, Pian X, Yin JL, Beck S, Lee N, Son J, Kim H, Yan C, Wang JH, Choi YJ, Whang KY. Opposite roles of MRF4 and MyoD in cell proliferation and myogenic differentiation. Biochem Biophys Res Commun 2007; 364(3): 476–482

[30] Sweetman D, Goljanek K, Rathjen T, Oustanina S, Braun T, Dalmay T, Münsterberg A. Specific requirements of MRFs for the expression of muscle specific microRNAs, miR-1, miR-206 and miR-133. Dev Biol 2008; 321(2): 491–499

[31] Perry SV. Vertebrate tropomyosin: distribution, properties and function. J Muscle Res Cell Motil 2001; 22(1): 5–49

[32] van de Meerakker JB, Christiaans I, Barnett P, Lekanne Deprez RH, Ilgun A, Mook OR, Mannens MM, Lam J, Wilde AA, Moorman AF, Postma AV. A novel α-tropomyosin mutation associates with dilated and non-compaction cardiomyopathy and diminishes actin binding. Biochim Biophys Acta 2013; 1833(4): 833–839

[33] Jones PA, Laird PW. Cancer epigenetics comes of age. Nat Genet 1999; 21(2): 163–167

[34] Boyd J, Risinger JI, Wiseman RW, Merrick BA, Selkirk JK, Barrett JC. Regulation of microfilament organization and anchorage-independent growth by tropomyosin 1. Proc Natl Acad Sci USA 1995; 92(25): 11534–11538

[35] Otto F, Kanegane H, Mundlos S. Mutations in the RUNX2 gene in patients with cleidocranial dysplasia. Hum Mutat 2002; 19(3): 209–216

[36] Lanaya H, Natarajan A, Komposch K, Li L, Amberg N, Chen L, Wculek SK, Hammer M, Zenz R, Peck-Radosavljevic M, Sieghart W, Trauner M, Wang H, Sibilia M. EGFR has a tumour-promoting role in liver macrophages during hepatocellular carcinoma formation. Nat Cell Biol 2014; 16(10): 972–981, 1–7

[37] Yannakoudakis BZ, Liu KJ. Common skeletal features in rare diseases: new links between ciliopathies and FGF-related syndromes. Rare Dis 2013; 1(1): e27109

[38] Vasikova A, Belickova M, Budinska E, Cermak J. A distinct expression of various gene subsets in CD34⁺ cells from patients with early and advanced myelodysplastic syndrome. Leuk Res 2010; 34(12): 1566–1572

[39] Lee K, Briehl MM, Mazar AP, Batinic-Haberle I, Reboucas JS, Glinsmann-Gibson B, Rimsza LM, Tome ME. The copper chelator ATN-224 induces peroxynitrite-dependent cell death in hematological malignancies. Free Radic Biol Med 2013; 60: 157–167

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg