Article Text

Download PDFPDF

Association between polymorphisms in microRNA seed region and warfarin stable dose
  1. Maryam Hosseindokht1,
  2. Hamed Zare2,
  3. Rasoul Salehi1,
  4. Leyla Pourgholi3,
  5. Shayan Ziaee3,
  6. Mohammadali Boroumand3,
  7. Mohammadreza Sharifi1
  1. 1 Department of Genetics and Molecular Biology, School of Medicine, Isfahan University of Medical Sciences, Isfahan, Islamic Republic of Iran
  2. 2 Cellular and Molecular Research Center, Birjand University of Medical Sciences, Birjand, Islamic Republic of Iran
  3. 3 Department of Pathology and Laboratory Medicine, Tehran Heart Center, Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran
  1. Correspondence to Dr Mohammadreza Sharifi, Department of Genetics and Molecular Biology, School of Medicine, Isfahan University of Medical Sciences, Isfahan 81744-176, Islamic Republic of Iran; mo_sharifi{at}; Dr Mohammadali Boroumand, Department of Pathology and Laboratory Medicine, Tehran Heart Center, Tehran University of Medical Sciences, Tehran, Islamic Republic of Iran; maboroumand{at}


Background The optimal dose of anticoagulant warfarin varies among patients to achieve the target international normalised ratio. Although genetic variations related to warfarin pharmacokinetics and vitamin K cycle are important factors associated with warfarin dose requirements, these variations do not completely explain the large interindividual variability observed in the most populations, suggesting that additional factors may contribute to this variability. microRNAs have recently been introduced as regulators of drug function genes, and therefore, may be involved in drug responses. In this study, we aimed to evaluate the possible association between variants in the seed region of microRNAs, which target the genes involved in the action of warfarin and warfarin dose requirement.

Methods 526 samples were collected from Iranian patients. Four selected polymorphisms in the seed region of microRNAs (rs2910164, rs66683138, rs12416605 and rs35770269 in miR-146a, miR-3622a, miR-938 and miR-449c, respectively) were genotyped by PCR-restriction fragment length polymorphism method.

Results rs2910164 C/G in the seed region of miR-146a was associated with warfarin dose requirement (p<0.001); the patients with GG genotype had the higher mean dose of warfarin (40.6 mg/week, compared with 33.9 and 31.8 mg/week for GC and CC genotypes, respectively). The association of other polymorphisms with warfarin dose requirement was not statistically significant.

Conclusion rs2910164 C/G in the seed region of miR-146a is associated with warfarin maintenance dose, likely by disrupting interaction between miR-146a and ATP-binding cassette subfamily B member 1 gene, ABCB1. Therefore, this polymorphism may possibly be a potential factor for assessment of warfarin dose requirements.

  • clinical pharmacology

Statistics from


Warfarin is the most common oral anticoagulation drug, prescribed for the prevention and treatment of thromboembolic disorders.1 However, given the variability of response to standard dose among individuals and narrow therapeutic index, choosing a correct dose of warfarin is remained challenging. The variability of response to standard dose of warfarin depends on several environmental and acquired factors including dietary intake, demographic factors, age, weight and concurrent medication.2 Furthermore, the response to warfarin is largely known to be determined by genetic background, that is, variations in the genes involved in the pharmacokinetics and pharmacodynamics of warfarin.2 3 The interindividual variability of response to warfarin means that a certain dose in an individual may result in rapid rise in international normalised ratio (INR), which in turn may lead to various complications including bleeding, while in other individuals, this dose may result in not achieving the demanded target range of INR, thus fail the therapy.4 Despite recent progresses in the development of new drugs, warfarin is still the first choice used for various diseases, including heart valve disease, and is one of the most commonly used drugs in the low-income and middle-income countries, highlighting the importance of prescription of the exact dose of this drug for the patients.5 It has been demonstrated that polymorphisms in CYP2C9 have profound effects in the metabolising activity of this enzyme and patients with some allelic variants may metabolise warfarin in a slower manner, which therefore require fewer dose of warfarin.6 However, allelic variants of cytochrome P450 enzymes, including CYP2C9, do not explain the large interindividual variability of warfarin observed in the most populations, suggesting that additional genetic factors may contribute to this variability.

Another important gene which, at least in part, contributes to the interindividual variability of response to warfarin, is the vitamin K epoxide reductase complex subunit 1 (VKORC1) gene, which codes a subunit of the vitamin K epoxide reductase enzyme.7 In this line, it has been reported that some single nucleotide polymorphisms (SNPs) could increase the effect of warfarin, resulting in an elevated tendency of bleeding in these subsets of population, while other SNPs may cause people suffer from warfarin resistance.8 However, both CYP2C9 and VKORC1 genetic variants account for no more than 35%–50% of the warfarin dose variability.9 There are also other genes involved in the intervariability of warfarin response. However, though helpful, the current genetic knowledge does not completely explain the warfarin dose requirements. Hence, there are much studies needed to unravel the genetic mechanisms underlying intervariability of warfarin dose.

Recent evidences have revealed that microRNAs may contribute to warfarin dose requirement and be associated with risk factors for arterial and venous thrombosis.10 Recent studies have indicated that targeting some microRNAs (miRNAs) may be useful to improve outcome of patients with venous thromboembolism.11 However, the studies on the association between polymorphisms in the seed region of miRNAs with warfarin dose requirement are almost lacking. In our previous study, we demonstrated the association between microRNA binding site-related SNPs and the incidence of warfarin-related bleeding.12 In the present study, we aimed to find out the possible association between miR-146a-3p rs2910164, miR-3622a rs66683138, miR-938 rs12416605 and miR-449c-3p rs35770269 variant polymorphisms and warfarin dose requirements in an Iranian population of patients with heart valve disease, deep vein thrombosis, atrial fibrillation and pulmonary thromboembolism.



This cross-sectional study was performed after approval of the local ethics committee. Thereafter, samples were collected from those who referred to Tehran Heart Center from January 2018 to August 2018. The inclusion criteria were as follow: (1) use of warfarin due to heart valve disease, deep vein thrombosis, atrial fibrillation and pulmonary thromboembolism (2) use of warfarin with stable therapeutic dose defined as the dose required to maintaining an INR in target range (prespecified as being in range of 2–3.5) in three consecutive blood tests. The exclusion criteria were as having kidney disease (blood creatinine level >2 mg/dL), hepatic disease, thyroid disease, cancer, acute infection, use of glucocorticoids and use of alcohol. A total of 526 patients met these criteria and were involved in the study.

Clinical information and data collection

Data on gender, daily warfarin dose, age, indication for therapy, warfarin therapy duration, body weight, INR measurements, height, comorbidity and concurrent medication were collected from the patient’s medical records. Data regarding the medication histories, diet, blood pressure, changes in warfarin dose and alcohol were collected at each physician visit. Variables such as cardiovascular risk factors and laboratory information were gathered by using hospital database. In this study, warfarin maintenance dose was defined as a dose of warfarin at which, INR is maintained at 2–3.5.

Biochemical analysis and prothrombin time-INR test

Three blood samples were collected from each subject and stored into three tubes; one tube for biochemical analysis (clot activator tube), one tube for prothrombin time (PT) test containing sodium citrate and another tube for molecular analysis (Ethylenediaminetetraacetic acid (EDTA) coated tube), which were stored at −20°C for DNA extraction.

Blood samples collected into the clot activator tube were sent to the Tehran Heart Center laboratory for biochemical tests and fasting blood glucose, creatinine, thyroid hormones (T3, T4 and thyroid stimulating hormone), and liver enzymes (alanine aminotransferase, aspartate aminotransferase and alkaline phosphatase) were measured.

For performing PT-INR tests, the plasma samples were separated from the cells within 30 min from the centrifugation and the samples were sent to Tehran Heart Center for PT-INR test using a Hemochron Signature Elite device. The PT responses were obtained in seconds, and then the INRs were calculated according to the international formula, INR=(PT Patient/PT Reference Plasma)ISI, while ISI denotes for International Sensitivity Index.

SNP selection

In this study, we used miRNA-SNP2.0 database for finding polymorphisms in the seed regions of miRNAs targeting genes involved in the action of warfarin. From the variants suggested by the database, we selected those with minor allele frequency >0.1 and with the highest difference in minimum free energy between the wild type and mutant alleles.

Table 1 shows the miRNA genes studied in this paper, their corresponding variant polymorphisms, putative target genes and the gain/loss status of the targets for each miRNA.

Table 1

The miRNAs, their corresponding variant polymorphisms and their putative targets

DNA extraction and analysis

For DNA extraction, 5 mL of whole blood was collected from each subject in an EDTA-coated tube (0.2 mL, 0.5 M) and DNA was extracted from the samples according to the standard salting-out protocol.13 Genotyping of rs2910164 G/C, rs66683138 A/G, rs12416605 T/C and rs35770269 A/T variant polymorphisms was performed using PCR-restriction fragment length polymorphism technique. Table 2 shows the primer sequences, fragments sizes and the enzyme used for the digestion of each polymorphism site. The PCR reactions were performed in the final volume of 25 µL and the reagents concentrations was as follow: 10×PCR buffer (Fermentas, Germany), 0.4 mM of each dNTP (TAKARA, Japan), 5 pmol of each primer (CinnaGen, Iran), 50 ng template DNA, 2 mM MgCl2 (Roche, Germany), 1 U Taq DNA polymerase (Takara, Japan) and sterile distilled water up to 25 µL. PCR amplifications were started with an initial denaturation step of 5 min at 94°C, followed by 30 cycles of 30 s denaturation at 94°C, 30 s annealing at 59°C for rs12416605°C and 60°C for other variant polymorphisms and 30 s extension at 72°C, finished by a final extension for 5 min (72°C).Thereafter, the PCR products were evaluated by subjecting to electrophoresis on 2% agarose gel, stained with GreenViewer and visualised by exposure to ultraviolet light. Then, the PCR products containing the rs2910164, rs66683138, rs12416605 and rs35770269 were digested by the restriction enzymes MnlI, MwoI and SmlI at 37°C, 60°C and 55°C overnight, respectively. The DNA fragments were analysed on 1.5% agarose gel electrophoresis.

Table 2

Primer sequences and restriction enzymes, used for each polymorphism

Statistical analysis

The genotype frequencies of the polymorphisms were analysed using a χ2 test, to examine that all polymorphisms were within Hardy-Weinberg equilibrium. Clinical and demographic characteristics were analysed in patients. Data are expressed as numbers (%) and mean±SD for categorical and numerical variables, respectively. Furthermore, one-way analysis of variance was performed to compare differences between the rs12416605 T/C, rs35770269 A/T, rs2910164 G/C and rs66683138 A/G genotypes and the warfarin dose using SPSS software (V.21). Throughout this study, p values less than 0.05 were considered as statistically significant.


In this study, of all the patients referred to Tehran Heart Center from January 2018 to August 2018, a total of 526 patients met the criteria for being included in the study. There were a total of 247 (47%) men and 279 (53%) women. The mean age was 59.87±12.52 years; the weekly warfarin dose was 36.05±17.77 mg and the duration of therapy was 1281 days (133–4432). The indications of warfarin therapy in this study were heart valve disease (333/526=63.3%), deep vein thrombosis (63/526=12%), atrial fibrillation (111/526=21.1%) and pulmonary thromboembolism (19/526=3.6%). The patients were subjected to different doses of warfarin and followed up to find their warfarin maintenance dose. Table 3 reveals the clinical and demographic characteristics of patients.

Table 3

Clinical and demographic characteristics of study population

The distribution of all polymorphisms was in Hardy-Weinberg equilibrium (χ2<3.84 and p>0.05). Table 4 demonstrates the genotype distribution and association of polymorphisms with warfarin dose in anti-coagulated patients. As shown in table 4, from the four polymorphisms evaluated in this study, only the polymorphism in miR-146a (rs2910164) had association with warfarin dose requirement (p<0.001); in patients with GG genotype, the mean warfarin dose for maintaining INR between 2 and 3.5 were 40.6 mg/week, whereas in those with GC and CC genotypes, mean doses of warfarin were 33.9 and 31.8 mg/week, respectively (figure 1 and table 4). The association between the other polymorphisms and warfarin dose requirement was not statistically significant (table 4).

Table 4

Genotypes frequency and association of polymorphisms with warfarin maintenance dose

Figure 1

Box plot graph showing the distribution of warfarin dose by rs2910164 genotype. boxes demonstrate the IQRs and median maintenance dose. The numbers above whiskers indicate mean values. Vertical lines below and above boxes demonstrate the minimum and maximum values.


The present study is a large series of patients with heart valve disease, deep vein thrombosis, atrial fibrillation and pulmonary thromboembolism and surveys the association between polymorphisms rs2910164, rs66683138, rs12416605 and rs35770269, located in genes coding for miR-146a-3p, miR-3622a, miR-938 and miR-449c-3p, respectively, and warfarin dose requirements. These miRNA polymorphisms have been previously reported to be associated with various diseases. For instance, the miR-146a rs2910164 variant polymorphism has previously been reported to be associated with an increased risk of venous thromboembolism,14 coronary artery disease15 and cancer.16

SNPs in miRNA genes have been associated with risk for various conditions, but the genotype distribution is population specific. Therefore, given the lack of studies on the role of miRNA polymorphisms in response to warfarin, this study was conducted in an Iranian population to appraise the association between the miRNAs polymorphisms, which their targets are related to warfarin effect and warfarin dose response.

According to this study, the only polymorphism which had association with warfarin dose requirement was rs2910164 in the gene coding for miR-146a. MiR-146a is a miRNA regulating the expression of multiple genes, including adenosine triphosphate-binding cassette subfamily B member 1gene, ABCB1. ABCB1 gene (also called MDR1, standing for multidrug resistance 1 gene) codes for a multidrug efflux pump, called P-glycoprotein (P-gp). P-gp is responsible for inhibiting the absorption of many drugs, including warfarin, as well as excreting them through kidneys and liver.17 Therefore, the level of warfarin in target cells may be dependent on this gp. In this regard, it has been demonstrated that polymorphisms in ABCB1 gene are associated with variability of dosage of warfarin.18 19 Given the regulatory effect of miRNAs on transcripts of human genes, polymorphisms in miR-146a are reported to be associated to conditions related to the ABCB1 gene.20 The possible reason for the effect of this variant (rs2910164) in the function of ABCB1 is that this single nucleotide variant is located within the seed region sequence of mature miR-146a, and consequently, affects the interaction of this miRNA with targets, resulting in alteration in the expression of ABCB1 transcript. In this regard, according to our observation, GG genotype of rs2910164 was associated with the higher mean dose of warfarin, suggesting that G allele in this locus may be corresponded to less effective targeting of ABCB1 by miR-146a, which justifies the need for higher dose of warfarin for the patients carrying this allele.

Previous studies have shown that SNPs in miR-146 may increase the susceptibility of cardiovascular disorders, including coronary artery disease. For instance, in a study by Wang et al, a total of 353 Chinese patients with coronary artery diseases and 368 controls were evaluated.21 According to this study, (C allele) of rs2910164 was more frequent in the patients. This report is inconsistent with our study which demonstrates that G allele may underlie warfarin resistance and the risk of disease. However, according to a meta-analysis by He et al, CC genotype of rs2910164 seems to be a protective factor for cardiocerebrovascular diseases in Chinese population.22 Furthermore, this study demonstrated that the contribution of rs2910164 variant alleles in cardiocerebrovascular diseases may be ethnicity specific; while CC genotype protects the Chinese patients for the disease, it increases the susceptibility of patients in Indian and Korean populations. Taken together, this meta-analysis demonstrated that rs2910164 CC genotype has a protective effect with overall cardiovascular diseases and the specific coronary artery disease, which is in consistent with our results. Studies on the effect of rs2910164 polymorphism on warfarin dose requirement in Iranian population are lacking. Some studies in Iranian population indicate that C allele of rs2910164 may be associated with coronary artery disease,23 while other studies indicate that C allele is associated with the risk of obesity.24 However, all these studies are performed in a limited number of patients and there are more studies and larger cohorts needed to evaluate the exact role of this polymorphism on cardiovascular diseases and warfarin dose requirements.

This study was a preliminary report of the effect of polymorphisms in the seed region of miRNAs on warfarin dose requirement in an Iranian population of patients with heart valve disease, deep vein thrombosis, atrial fibrillation and pulmonary thromboembolism. The study benefits from a large cohort of Iranian patients and demonstrates the importance of rs2910164 SNP in the seed region of miR-146a. However, given the limited number of miRNAs studied in this work, the evaluation of a larger list of miRNAs will help unravel more detailed information with regard to miRNA polymorphisms and warfarin dose requirements.


Our study suggests that rs2910164 in the seed region of miR-146a is an independent factor associated with warfarin dose requirement in patients with heart valve disease, deep vein thrombosis, atrial fibrillation and pulmonary thromboembolism. This association may be due to the effect of this SNP on interaction of miR-146a with ABCB1 mRNA, which in turn could alter the expression of ABCB1 and change in the absorption of warfarin to the cells. The other polymorphism variants show no association with warfarin dose requirement.

Main messages

  • rs2910164 in miR-146a is related to the warfarin stable dose.

  • miR-3622a, miR-938 and miR-449c polymorphisms were not associated with warfarin dose.

Current research questions

  • More studies are needed to confirm the exact role ofrs2910164 and evaluation of the other SNPs in a larger cohort of Iranian population.

  • Transcriptomic of studied miRNAs and their targetmRNAs/proteins should be analyzed.

What is already known on the subject

  • Variations in vitamin K epoxide reductase complex subunit 1 and CYP2C9 do not completely explain the interindividual variability in warfarin dose requirement.

  • microRNAs may contribute to warfarin dose requirement.



  • Contributors MH, MB and MS: designed and performed experiments, analysed data and cowrote the paper. LP, HZ and SZ: performed experiments. HZ: performed bioinformatic analyses. MS, MB and RS: supervised the research.

  • Funding This study was supported by Isfahan University of Medical Sciences (grant number: 396319).

  • Competing interests None declared.

  • Patient consent for publication Not required.

  • Ethics approval This study was approved by the local ethics committee of Isfahan University of Medical Sciences, IRAN. The studies have been approved by the appropriate institutional and/or a national research ethics committee and have been performed, in accordance with the ethical standards, as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data availability statement Data are available on reasonable request.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.