Reach Us +44-1904-929220
Using microRNA as Biomarkers of Drug-Induced Liver Injury | OMICS International
Journal of Molecular Biomarkers & Diagnosis

Like us on:

Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on Medical, Pharma, Engineering, Science, Technology and Business
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Using microRNA as Biomarkers of Drug-Induced Liver Injury


miRNA; Drug-induced liver injury; Biomarker


MicroRNAs (miRNA) are small non-coding RNAs that regulate gene expression post-transcriptionally [1]. It has been estimated that >60% of human protein-coding genes are regulated by miRNAs [2]. They typically down regulate gene expression; however, the roles of miRNA are still evolving and some miRNA have been shown to increase specific gene translation [3]. miRNA have even been shown to be released from cells and regulate gene translation in distant cells, reminiscent of cell-to-cell communication by cytokines and other soluble factors that are actively released from cells [4,5]. To date, over 720, 408, and 1424 miRNAs have been identified in mice, rats, and humans, respectively (Sanger miRBase Release 17).

miRNA holds a unique position among RNA for use as a biomarker due to its unique stability. Unlike mRNA, miRNA has been shown to be remarkably stable in a variety of tissues and body fluids [6-8].This stability greatly facilitates its use as a clinical biomarker of disease and injury since sample handling and processing are much less problematic when compared to mRNA. miRNA-based biomarkers also have many advantages over protein-based biomarkers primarily due to the fact that miRNA is a relatively simple molecule that can be detected using standard, robust molecular biology techniques such as real time quantitative polymerase chain reaction (RT-qPCR).

miRNA expression profiles have been extensively investigated for distinguishing cancerous vs. non-cancerous tissue [1,9-12].Profiles of miRNA in cell-free body fluids have also been able to distinguish patients with different types of cancer and even provide prognostic information about disease outcome [13,14]. The hypothesis is that cancerous masses release miRNA into the systemic circulation and, therefore, changes in the pattern and amount of miRNA can be used to detect the type of cancer. Studies are showing promise in using miRNA in cell-free body fluids to detect organ injury. Since some miRNA exhibit tissue specific expression, it is possible that miRNA profiles could be used to not only assess tissue injury but also distinguish between different organs (e.g., heart vs. liver). The following sections provide an overview of the role of miRNA in the cell, review the relevant literature on using miRNA profiles to identify organ injury, and highlight the use of miRNA in assessing drug-induced liver injury (DILI).

miRNA transcription, processing, and function

miRNAs are a newly discovered class of non-coding RNA that regulate entire intracellular pathways primarily at a posttranscriptional level. miRNAs are approximately 21 - 24 nucleotides long and are highly conserved among species. Some miRNAs are derived from introns of protein-coding genes; whereas, others are derived from non-proteincoding RNA transcripts. Both of these miRNA derivation processes have provided insight into the unique role for these regions of DNA that were previously believed to have no relevant cellular function. Many miRNAs are located in clusters of 2-19 miRNA hairpins encoded in close proximity and can be derived from a single transcript. The processing of miRNA into the final mature form differs from mRNA and involves transcription from the relevant sequence of DNA and subsequent unique cleavage steps. The sequence of events is diagramed in Figure 1 and more details can be found in various reviews of the synthesis, processing, and function of miRNA [15-19].

Similar to mRNA, most miRNAs are transcribed by RNA polymerase II; in this case, primary miRNAs (pri-miRNAs) are generated. Pri-miRNAs are produced in a wide array of lengths, often several thousand nucleotides long, and similar to mRNA are capped at the 5’end and polyadenylated at the 3’ end [15,18]. Within the nucleus, the pri-miRNAs are processed by the ribonuclease III enzyme Drosha in cooperation with DGCR8 releasing small, approximately 70-nucleotide-long, stem-loop-structured molecules called precursor miRNAs (pre-miRNAs). The pre-miRNAs are exported from the nucleus into the cytoplasm by the protein Exporten 5 in a GTPdependent process. Once in the cytoplasm, the pre-miRNAs are processed into approximately 22-nucleotide long mature RNA duplexes by the enzyme Dicer in cooperation with the double-stranded RNA-binding protein TRBP. The mature miRNA duplex consists of a guide strand and a passenger strand. The Dicer-TRBP complex, in cooperation with argonaute 2, unwinds the duplex [20]. The guide strand is then incorporated into the argonaute-containing RNA interference-induced silencing complex (RISC) while the passenger strand is degraded [21]. The exact mechanism of strand selection is unclear but the guide strand generally exhibits lower thermodynamic stability at the 5’ end [22].

The miRNA/RISC complex is the ultimate effector of miRNAmediated gene repression. Most reactions involve binding of the miRNA/RISC complex to the 3’ untranslated region of the target mRNA; however, the mechanisms of miRNA regulation of gene expression are still evolving and there are likely other mechanisms that result in gene regulation such as binding to the 5’ untranslated region of the target mRNA [9]. If the miRNA exhibits perfect complementation to the target mRNA sequence, the mRNA is targeted for degradation. If the miRNA has less than perfect complementation, mRNA translation is repressed. In mammalian cells, the seed region of miRNAs (2-8 nucleotides) is the primary determinant of target recognition within the 3’-untranslated region of the target mRNA [23]. The majority of reactions requires perfect complementation between the seed region and the target mRNA; whereas, pairing outside the seed region is less stringent. Since the seed region is small and it is the primary target recognition determinant, a single miRNA can potentially regulate hundreds of target mRNAs. The concentration of a given miRNA can be regulated at all of the various processing steps, providing additional control over gene regulation [16].

Many studies have shown that miRNAs are deregulated under different pathological conditions, such as cancer and liver injury [10,15,17,24]. In the case of cancer, up-regulation of miRNA expression can induce or suppress tumor formation depending on their normal role of inhibiting tumor suppressive or oncogenic target mRNAs, respectively. For example, miRNA-21 targets multiple tumor suppressor genes, including Bcl2, PTEN, and Fas ligand. Increased expression of miRNA-21, as found in some cancers, results in the downregulation of multiple mRNA targets and subsequent increased proliferation and/or decreased apoptosis [16]. In general, miRNAs and their targets have reciprocal expression patterns and miRNA regulation of gene expression appears to act like a rheostat to fine tune gene expression of many genes as opposed to gross regulation of a single gene [25]. In addition, there is redundancy built into miRNA regulation of gene expression with many miRNAs being part of families with related sequences that are potentially involved in the same pathways [15]. Overall, miRNAs play a pivotal role in regulating gene expression and changes in miRNA expression are likely to either alter biological pathways, such as with cancer induction, or represent the response to changes in cellular homeostasis, such as during DILI. From a biomarker perspective, both of these factors are important and changes in miRNA expression are likely to serve both as biomarkers of disease/injury and provide novel insights into the biological pathways that are altered during disease/injury.

DILI is one of the leading causes of drug attrition during drug development and post-marketing drug withdrawal [26-29]. DILI is also a common cause of patient morbidity. Current biomarkers can detect liver injury but there are many inadequacies that make them less than ideal. For example, serum levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) are the primary biomarkers of hepatocellular injury. Unfortunately, limitations exist with the use of both enzymes since 1) elevations can reflect nonhepatic injury, particularly skeletal muscle injury, 2) elevations can occur after a critical therapeutic window has passed, such as with acetaminophen overdose, 3) elevations can occur in the absence of histological evidence of injury, such as with statins, 4) concentrations do not discriminate between different etiologies of liver injury, and 5) levels do not provide any insight into disease prognosis. The histopathological analysis of liver biopsies is another marker for liver disease; however, the biopsies are invasive and not routinely performed. Therefore, there is a need for new biomarkers that are not just sensitive but are also specific and prognostic.

The liver is anatomically and physiologically complex and liver injury can affect one or more of these factors. Therefore, DILI presents as diverse biochemical, histologic, and clinical abnormalities and can mimic all forms of liver injury ranging from asymptomatic elevations of hepatic enzymes to fulminant hepatic failure [30]. With the advent of various “omics” technologies (i.e., genomics, proteomics, and metabolomics), the spectrum of hepatotoxic effects caused by different drugs has widened exponentially such that it is unlikely that any two drugs will produce the exact same spectrum of effects using these exquisitely sensitive techniques that involve thousands of different endpoints [31]. However, even though the spectra are unlikely to match exactly, there are likely to be sufficient similarities to group drugs into broad categories [32].

Clinically, DILI is classified in terms of the clinical liver disease; the three most common types of injury are hepatocellular, cholestatic, or mixed hepatocellular/cholestatic. Hepatocellular often involves direct damage of the hepatocytes such as the centrilobular necrosis caused by acetaminophen. This type of damage is often associated with elevated serum ALT and AST levels due to leakage from damaged hepatocytes. Cholestatic injury often involves damage to some part of the bile excretion apparatus resulting in impaired bile excretion. This type of injury is often associated with elevated serum levels of biomarkers reflective of bile duct injury such as gamma glutamyl peptidase, alkaline phosphatase, and bilirubin. Mixed injury presents with a mixture of both types of effects. In addition to these three clinical types of injury, each one may be associated with a systemic syndrome. For example, in addition to the liver injury, some drugs may cause hypersensitivity reactions in other organ systems (e.g., Stevens-Johnson syndrome, renal dysfunction, myocarditis) suggesting an immunoallergic component to the injury [33,34]. Unfortunately, drugs rarely produce a single clear clinical picture making the diagnosis of DILI difficult. For example, amoxicillin/clavulanic acid usually causes cholestatic injury but can also produce acute hepatocellular injury or a mixed type injury [30,35]. There are also other types of DILI such as microvesicular steatosis, non-alcoholic steatohepatitis, chronic hepatitis, cirrhosis, and venoocclusive disease; however, these are much less common.

Despite the broad classification of drugs as causing hepatocellular, cholestatic, or mixed injury, they all are likely to share the common trait of having a mechanistic basis to the injury. For a compound like acetaminophen, the biochemical basis has been studied for years and has explained the clinical picture of injury [36]. However, even for this extensively studied drug, research continues to fine tune our understanding. The mechanism of injury for most compounds with a very confusing clinical picture (e.g., mixed injury with systemic hypersensitivity reaction) is unknown but it is likely that there is a biochemical basis behind the injury. The adaptation seen to the DILI caused by some drugs provides clues about the molecular processes that might play a role in DILI. For most patients on a given drug, no adverse effects are noted; however, some patients may experience transient, asymptomatic increases in serum ALT. The majority of these patients adapt to the drug as indicated by a return of serum ALT to baseline levels despite continued treatment. A small percentage of these patients will continue to experience increased ALT levels evolving into other clinical signs of liver injury and possibly leading to liver failure. Although the clinical outcome is much different for the three groups of patients (non-responders, adapters, and sensitive), the basic biochemical events are likely to be similar. The difference separating the three groups of people could be due to a multitude of reasons and include both genetic and environmental factors [37]. Since the cellular response to the drug involves a common biochemical basis, it is likely that “omics” analyses, such as miRNA analysis, will reveal the affected pathways and can be used to assess not only the type of liver injury but the mechanisms behind the injury.

Using miRNA profiles to detect organ injury

Starting in the late 1990’s, a wide array of studies have shown that miRNA levels are differentially expressed in the target tissue during disease or injury. It is important to note that patterns found to date have not been subjected to rigorous qualification so remain unproven. Cancer has been the focus of much study and it has been clearly shown that miRNA expression patterns differ between cancerous and non-cancerous tissue [10,38-43]. More recently, miRNA expression profiles after tissue injury have been able to distinguish normal versus injured tissue (reviewed below). These studies have clearly shown that miRNA profiles are altered in the target tissue during disease or injury; however, their application to clinical diagnosis is limited since they require invasive sampling of the tissue.

The ideal DILI biomarker would be present in readily accessible body fluids such as blood or urine and would meet the criteria listed in Table 1. Many studies have shown that cell-free miRNA is present in readily accessible body fluids and meet many of the ideal biomarker criteria (reviewed below). Studies have shown that the level of plasma or serum miRNA is altered in patients with various types of cancer [14,44-51]. In some diseases, miRNA levels are increased and in others they are decreased, but they share the common feature of being able to distinguish normal patients from those with cancer. For example, four serum miRNAs were differentially expressed in patients with hepatocellular carcinoma and the level of miRNA-221 correlated with tumor size, cirrhosis, and tumor stage [52].

Although miRNA detection is theoretically simpler and subject to fewer confounders than developing protein based assays, there are some technical issues associated with miRNA measurement. There are several different platforms for miRNA detection including Northern blotting, RT-qPCR, microarray, and Next-Generation sequencing. Unfortunately, at this time, there appears to be limited correlation between the different platforms, most likely due to the strong influence of different primer designs on the measurements for RT-qPCR and microarrays, the two most commonly used detection methods [53]. Many different protocols for sample preparation are used making comparisons between studies difficult. In contrast to total RNA, assessment of miRNA quantity and quality is less than ideal. For sample analysis, some studies use an enriched small RNA fraction whereas others use total RNA; however, the influence of the fractionation procedure has not been systematically investigated. Finally, and especially for cell-free miRNA analysis, a variety of different normalization processes have been used to control for technical and biological variability but none of them are ideal [14,39,53,54].

miRNA levels in the plasma or serum have been used to distinguish patients suffering from different types of organ injury, such as myocardial damage. A multitude of studies have shown that plasma/ serum miRNA levels can distinguish normal patients or animals from those experiencing myocardial injury [55-59]. For example, miR-1 was significantly elevated in the plasma from patients suffering acute myocardial infarction compared to controls [60] and, in a separate study, a panel of 20 miRNAs was shown to predict acute myocardial infarction in patients with high specificity, sensitivity, and accuracy although these have not been put forward for biological qualification [61]. Studies have also shown that the plasma/serum miRNAs are altered for a variety of other types of organ injury or disease such as cerebral infarction, skeletal muscle damage, muscular dystrophy, systemic lupus erythematosus, traumatic brain injury, and diabetes [62-67]. Several recent studies have shown that the plasma/serum level of various miRNAs are altered after various types of liver injury and disease such as hepatitis B [68] and C [69] infection, chronic hepatitis [70], non-alcoholic fatty liver disease [70], and hepatic injury and rejection after transplantation [71]. These studies clearly show that miRNAs are differentially released during organ injury and can be used to identify affected patients. Figure 2 diagrams the likely progression of miRNA release into the blood that accompanies organ injury, in this case, hepatocellular injury. miRNAs present within circulating blood cells are also being explored to detect organ injury. A recent study showed that miR-1291 and miR-663b in the peripheral total blood are highly specific and sensitive biomarkers to determine acute myocardial infarction [61]. In rat models of brain injury, miR-298, miR-155, and miR-362-3p in the blood have been proposed as candidate biomarkers to predict such damages [72]. In kidney disease, blood miR-142-5p, -155, and -223 have been shown be to up-regulated and correlated well with the human renal allograft status [73]. These studies indicate that circulating blood cells and their miRNA levels may act as sentinels of disease in other organs.

In normal cells, miRNA are released into extracellular fluid, such as blood, via several different processes as outlined in Figures 1 and 2; however, it is important to keep in mind that this is an evolving area of research and there are likely additional mechanisms for extracellular miRNA release that have yet to be discovered [74]. The first two mechanisms involve release of miRNA in cell-derived vesicles. Microparticles are vesicles that are derived from outward budding of the cell membrane and contain various cellular constituents, including miRNA. They are typically 100-1000 nm in size. Exosomes are internally derived vesicles that are contained within multivesicular bodies, the cellular organelles that integrate both endocytic and secretory pathways. Some multivesicular bodies fuse with the plasma membrane and release their contents (i.e., the exosomes) into the extracellular fluid. Exosomes are typically 30-100 nm in size thus being smaller than microparticles. The third mechanism involves release of miRNA bound to cellular proteins, such as argounate-2 and HDL [75,76]. Various studies have shown that these extracellular miRNAs can alter biological processes in distant cells and may play a role in cellto- cell communication [5,77-80].


Figure 1: miRNA transcription, processing, gene targeting, and extracellular release. After processing and selection of the guide strand, miRNA may be released from the cell as 1) microparticles, 2) exosomes from microvesicular bodies, and 3) & 4) protein complexes. RISC is the RNA interference-induced silencing complex that contains argonaute proteins.


Figure 2: Pathways of miRNA release from hepatocytes after exposure to a hepatotoxic drug. The cellular level of a given miRNA may be increased or decreased leading to corresponding changes in the amount of miRNA released from the cell. miRNA release processes may be altered such as A) microparticles, B) exosomes, and C) protein complexes. Apoptotic or necrotic cell death compromises the cell membrane causing leakage of miRNA.

In cells undergoing stress, homeostasis is altered leading to altered blood miRNA levels via a variety of different mechanisms (Figure 2). Cells undergoing stress may alter the cellular levels of miRNAs and this in turn leads to altered blood levels without any change in the normal processes of miRNA release from the cell. Disturbances in the release processes due to cell stress would directly influence how much miRNA is released into the blood. If cell injury is severe and apoptotic or necrotic death ensues, it is likely that in addition to changes in the normal synthesis and release pathways, miRNAs are also released through the compromised cellular membrane. The interplay of these factors will ultimately determine how the concentration of a given miRNA in the blood changes after tissue injury and in what form the miRNA is released (e.g., microparticle, exosome, or protein complex).

The alteration of cell-free miRNAs is not restricted to the blood. Altered levels of miRNAs in a wide array of body fluids such as urine, sputum, feces, bile, cerebrospinal fluid, and saliva have been detected in patients suffering various diseases or organ injury [8,77,81-86]. Similar to blood, miRNA in these body fluids appear to be much more stable than exogenously added RNA. It is clear that cell-free miRNAs are present in a wide array of body fluids and they hold great promise in the diagnosis of organ injury.

miRNA as biomarkers of DILI

As mentioned previously, new biomarkers of DILI are needed since the current biomarkers have various flaws such as induction by other types of organ injury or lack of prognostic value. As shown for a variety of different types of diseases and organ injury, miRNA shows great promise of meeting many of the criteria of an ideal DILI biomarker (Table 1). For the liver, a variety of different diseases and injury such as acetaminophen- and carbon tetrachloride-induced hepatotoxicity, nonalcoholic steatohepatitis, FAS-induced acute liver injury, ischemia/ reperfusion, and fibrosis have been shown to alter hepatic miRNA levels [87-93]. In a seminal study by Wang et al. [94] that used an acetaminophen-induced mouse model of hepatotoxicity, the level of many plasma miRNAs inversely correlated with the level of hepatic miRNAs, indicating that for these miRNAs, hepatic injury caused the release of miRNAs into the circulation. For example, miRNA-122 and miRNA-192 exhibited high levels in the liver and upon injury were released into the blood with concurrent decreases in the liver. The changes in these miRNAs not only preceded changes in serum ALT levels but also exhibited much less variability in animal to animal responses. Similar clinical results were recently obtained in patients with acetaminophen poisoning [95]. The serum level of two relatively liver-specific miRNAs, miR-122 and miR-192, were substantially higher in patients suffering from acetaminophen-induced acute liver injury compared to healthy controls. The serum level of a heartenriched miRNA, miR-1, was not altered in these patients and the serum level of a brain-enriched miRNA, miR-218, was slightly higher in the acetaminophen patients. Serum miR-122 levels correlated with peak ALT but not prothrombin time or total bilirubin concentrations. In a smaller cohort of patients suffering from non-acetaminopheninduced acute liver injury, serum miR-122 was also increased. Other studies have also shown that plasma or serum miRNAs are altered after other types of liver injury and disease [69-71,96-98].


• Preferentially (or exclusively) produced in target tissue
• Able to differentiate pathologies


• Early indication of disease before clinical symptoms develop
• Present at low concentrations in controls and exhibits significant increase after injury
• Long half-life


• Proportional to degree of severity of pathology


• Can predict disease progression


• Not confounded by unrelated conditions
• Rapid, simple, accurate, and inexpensive detection


• Data can be used to bridge pre-clinical and clinical results


• Released from tissue into accessible fluid sample

Table 1: Characteristics of an ideal DILI biomarker.

We have recently shown that urinary miRNA levels can distinguish control from acetaminophen-treated rats [99]. Similar to the plasma miRNA results from Wang et al. [94], urinary miRNA levels were both more sensitive and more consistent biomarkers of acetaminophen exposure when compared to serum ALT and AST. For example, urinary miRNAs were increased at a low acetaminophen dose that failed to alter serum ALT and AST. In addition, at the high acetaminophen dose, only 2 of 7 animals exhibited increased ALT/AST levels; whereas, every animal had elevated urinary miRNA concentrations for at least several miRNAs that were screened. These results show that urinary miRNAs may be suitable biomarkers for liver injury; however, additional work is need to determine the specificity of the findings and if they are translatable to the clinic.

Qualification of cell-Free miRNA as biomarkers of DILI and data need

The available published data show that the level of various miRNAs in serum, plasma, and/or urine may serve as new biomarkers of DILI. However, more studies and data are required before the various miRNAs are qualified as true biomarkers of DILI that can supplement and/or replace existing biomarkers. The FDA and ICH have issued guidance on biomarker qualification and the content of data submissions [100- 102]. Although specific testing and qualification plans are not provided in the guidance documents, they do highlight the fact that robust data are required to qualify a new biomarker. As part of the qualification process, it will be important to determine how cell-free miRNAs meet the various specifications of an ideal biomarker (Table 1). Although cell-free miRNAs do not have to exceed every criterion, it will be necessary to determine how they compare to existing biomarkers of DILI. It will also be important to determine if individual or batteries of cell-free miRNAs are associated with specific drugs, diseases, or injuries or if they are just a generic indicator of gross liver injury, regardless of the insult. Inter-individual differences and variability in responses will need to be determined. Most marketed drugs that cause DILI fall into the rare “idiosyncratic” category and adversely affect only a limited number of patients. Ideally, cell-free miRNA will be shown to be superior to existing biomarkers in identifying idiosyncratic DILI earlier in the course of the disease so that the drug can be withdrawn before serious injury occurs. Finally, no existing biomarker of DILI provides prognostic information and it will be important to determine if cell-free miRNA fills this data gap.


Over the past decade and especially in the past few years, a lot of research has been conducted looking at the ability of miRNAs to serve as biomarkers of disease or injury. miRNAs hold great promise as biomarkers in extracellular fluids since they are stable, have a simple chemistry, are potentially organ specific, and are easily detected using readily available and sensitive methods. The use of cell-free miRNAs as biomarkers of DILI is in its infancy but several studies have shown promising results and cell-free miRNAs have been successfully used to detect other types of organ injury. Future studies will help determine the role cell-free miRNAs play in detecting DILI and whether they will simply be adjuncts to existing biomarkers or will show superior performance and supplant existing biomarkers.


Dr. Xi Yang is supported by the Research Participation Program at the National Center for Toxicological Research administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the US Food and Drug Administration.

The opinions expressed in this manuscript do not reflect the official positions or policies of the U.S. Food and Drug Administration.


Visit for more related articles at Journal of Molecular Biomarkers & Diagnosis


MicroRNAs (miRNA) are small non-coding RNAs that regulate gene expression post-transcriptionally. They bind to complementary sequences on target mRNA and typically down regulate expression or increase the rate of degradation; however, the roles of miRNA are still evolving and some miRNA have been shown to increase specific gene translation. miRNA holds a unique position among RNA for use as a biomarker due to its unique stability. Unlike mRNA, miRNA has been shown to be remarkably stable in a variety of tissues and body fluids. This greatly facilitates the use of miRNAs as clinical biomarkers of disease and injury since sample handling and processing is much less problematic when compared to mRNA. miRNA expression profiles have been extensively investigated for distinguishing cancerous vs. non-cancerous tissue. Taking this approach one step further, profiles of miRNA in cell-free body fluids have also been able to distinguish patients with different types of cancer and even provide prognostic information about disease outcome. The rationale behind this approach is that cancerous masses release miRNA into the systemic circulation and changes in the pattern and amount of miRNA can be used to detect the type of cancer. A recent extension of this approach is using miRNA in cell-free body fluids to detect organ injury. Several studies have shown increased serum levels of specific miRNA after myocardial or hepatocellular injury. Since some miRNA exhibit tissue specific expression, it is possible that miRNA profiles could be used to not only identify gross organ injury but also distinguish between different types of organ injury (e.g., heart vs. liver). This article will provide an overview of the role of miRNA in the cell, review the literature on using miRNA profiles to identify organ injury, and highlight the potential use of miRNA for assessing drug-induced liver injury. It should be noted that at the time of this writing, none of the profiles have been qualified for clinical use by the FDA.


Asthma; COPD; CD40 gene polymorphism; CD40 expression


Asthma and chronic obstructive pulmonary disease (COPD) show similarities and substantial differences. It is stipulated that asthma and COPD have common genetic and environmental risk factors, which ultimately lead to clinical disease depending on the timing and type of environmental exposures. Thus, a particular group of shared genetic factors may lead to asthma when combined with specific environmental factors, whereas combination with other environmental factors, will lead toward COPD [1].

CD40 is a member of the tumor necrosis factor receptor superfamily, that plays a substantial, multi-faceted role in inflammation [2]. Previous studies have shown that interactions between CD40 and its ligand CD154 (CD40L) have been implicated in lung disorders. CD40 which is found on a variety of inflammatory cells, induces the release of inflammatory mediators and plays a role in airway inflammatory responses [3].

Chronic inflammation of the airways plays a major role in the pathogenesis of asthma as well as chronic obstructive pulmonary disease (COPD). Their pathogenesis is influenced by both environmental and genetic factors [4,5].And since CD40 signaling has been linked to pathogenic processes of chronic inflammatory diseases, [2,3] therefore CD40 polymorphism could be associated with these two diseases.

Until now, only little information had explored the association between CD40 polymorphisms and genetic susceptibility to airway inflammatory disease. A C/T in the 5’ untranslated region of CD40 located at the −1 position within the Kozak sequence (rs1883832) has been associated with CD40 protein expression [6].Therefore, we hypothesized that the CD40 gene (− 1C/T) polymorphism may have a role in the genetic susceptibility to asthma and COPD. Therefore, we chose to study CD40 gene (−1C/T) polymorphism and investigate its functional effect on the expression of CD40 in a genetic study of asthma and COPD in the Egyptian population.

Patients and methods

Our study included 150 subjects who were divided into three groups :

group 1 included 50 asthmatic patients,

group 2 included 40 COPD patients, and

group 3 included 60 normal subjects as control.

Asthmatic patients were diagnosed according to the standard criteria, as previously described
[ 7]. Asthmatic patients exhibited airway reversibility, as defined by an inhalant bronchodilator-induced improvement of forced expiratory volume in one second (FEV1) by more than 12% or 200 ml.

The diagnosis of COPD was based on the definition provided by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) [5]. As defined by a post-bronchodilator forced expiratory volume in one second (FEV1) to forced vital capacity (FVC) ratio of < 70% and a β2-agonist reversibility of <12% or 200 mL.

Patients were excluded from the study if they had other respiratory diseases, diabetes mellitus, cardiac diseases or thyroid disease.

The control group included normal subjects who were recruited from the general population who had no respiratory symptoms, and no evidence of airflow obstruction. The majority of them were either current or ex-smokers. Individuals were excluded if they had a history of chronic lung disease or atopy, an acute pulmonary infection in the 4 weeks preceding assessment for the study, or a family history of asthma or COPD.

All the cases and controls were unrelated Egyptian people who were selected from the same population. The case groups were recruited from Alexandria main university hospital. All subjects were enrolled in the study after a written informed consent according to the protocol approved by the Ethics Committee of the Alexandria Main University Hospital. Peripheral venous blood samples of 5 ml were drawn from each individual by standard venepuncture. The blood sample was divided into two aliquots; one in a sterile tube with K2- EDTA anticoagulants for flow cytometry and genotyping, and the other one was collected in a plane tube for serum separation.

Analysis of -1C/T SNP of the CD40 gene (rs1883832)

Total genomic DNA was extracted from whole blood samples using QIAamp DNA Blood Extraction Kit (Qiagen,UK). The CD40- 1C/T (ref SNP ID: rs1883832) polymorphism was determined by using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) analysis. According to Hsieh et al. [8] the primer sequences were: Forward, 5’-TACACAGCAAGATGCGTCC CT-3’; Reverse, 5’-AACAACTCACAGCGGTCAGCAA-3’. A total of 50 ng genomic DNA was mixed with 0.25 μl primers in a total volume of 25 μl containing 10 mM Tris-HCL pH 8.3, 50 mM potassium chloride, 2.0 mM magnesium chloride, 3 μl each of dNTP, and 1 U Taq DNA polymerase. The PCR amplification was performed in a programmable thermal cycler with initial denaturation step for 5 min at 95ºC, 35cycles of denaturation for 30s at 95ºC, annealing for 30s at 65ºC and extension for 30s at 72ºC, followed by a final extension step for 10 min at 72ºC.

The amplified products were digested by 5U of restriction enzyme NcoI for 8 h, according to the manufacturer’s recommendation (Fermentas, Burlington, Ont., Canada). NcoI digestion cleaves the 310 bp PCR products into 2 fragments of 249 and 61bp when C allele is present. The PCR products in 0.2 ml, 12-tube strips, were transferred directly from the thermocycler into the sample tray of the QIAxcel Capillary electrophoresis from Qiagen company [9]. Separation was performed in a 12-channel gel cartridge (GCK5000) purchased from eGene Inc. (Qiagen, USA). The sizes of the alleles resolved from the subsequent separation were automati-cally calculated in bp and exported using the BioCalculatorTM software, which provides a gel view and an electro-pherogram of the separation. QIAxcel DNA High Resolution Kit was used with Alignment Marker15-1000bp and size Marker 50-800bp.

Dual expression of CD20 and CD40 using flow cytometry: As previously described [10,11] B-cell lineage lymphocytes were stained with a phycoerythrin (PE)-conjugated anti-CD20 mouse monoclonal antibody and a fluorescein isothiocyanate(FITC)–conjugated anti- CD40 mouse monoclonal antibody (both from BD Biosciences, San Diego, CA). Gating was done on CD20 positive population of B-lymphocytes and were further analysed for staining with CD40 (Dual color). Phycoerythrin- and fluorescein isothiocyanate–conjugated mouse IgG1k antibodies (BD Biosciences) were used as isotypematched negative controls. Dual expression of CD20 and CD40 cells was analyzed by flow cytometry : Facscalibur from BD .

Determination of Immunoglobulin E (Ig E) concentration: Ig E concentration in human serum was detected by electrochemiluminescence on Cobas E411 from Roche (Germany) [12].It was determined in the serum of asthmatics and controls.

Statistical analysis

Statistical analysis of data was performed using the PASW version 18. Genotype and allele frequencies of CD40-1C/T were in agreement with Hardy-Weinberg equilibrium. Genotype frequencies were compared among the studied groups using the chi square test and fisher’s exact test when appropriate. Odds ratio (OR) and 95% confidence intervals (CI) were calculated to assess the relative risk conferred by a particular allele or genotype. Demographic and clinical data between groups were compared by the chi square test and by the Student’s t-test. Logistic regression analysis controlling for age, sex and smoking status as covariates was done to estimate the risk in asthma. Only the smoking status was used as a covariate in the logistic regression analysis done to estimate the risk in COPD. Mann Whitney test was done for the comparison of CD40 expression according to different genotypes in the 3 groups. Statistical significance was assumed at P ≤0.05.


Characteristics of the study population

The characteristics of the study population are summarized in Table 1. Age sex, smoking status and the total serum IgE showed a significant difference between asthmatic patients and the control group. However COPD patients showed a significant difference only in the smoking status when compared to controls.

  Asthmatic patients COPD patients Control subjects p1 p2
Subjects (n) 50 40 60    
Age (years) 38.500±10.53 59.350 ± 10.39 52.150±16.89 <0.001* 0.414
Sex M/F (%) 25(50%)/25(50%) 30(75%)/20(25%) 45(75%)/1525% 0.007* 1.000
Smoking status n(%)          
Non smoker 33 (66%) 7 (17.5%) 19 (31.7%) <0.001* <0.001*
Current smoker 10 (20%) 19 (47.5%) 30 (60%)
Ex smoker 7 (14%) 14 (35%) 11 (22%)
Pack-year of smoking 15.54 ± 7.43 31.60 ± 13.81 23.93 ± 14.09 0.038* 0.024*
Post –BD FEV1 % predicted 74.78 ± 9.29 66.8 ± 12.93 - - -
Spirometric classification of severity Intermittent= 7(14%)
Mild= 18 (9%)
Moderate= 28 (56%)
Severe= 6 (12%)
Mild= 10(25%)
Moderate= 21(53%)
Severe= 3(8%)
Very severe= 6(15%)
Total IgE, IU/ml 221.11 ± 75.51 - 38.40 ± 12.89 <0.001* -

Table 1: Characteristics of the study population.

Distribution of genotypes and alleles of CD40-1C/T polymorphism: The genotype frequencies were in agreement with the Hardy–Weinberg equilibrium in control group. Genotype distribution showed different patterns among the three studied groups (Figure 1). Where the CT genotype was prevailing in the COPD patients and control group. While in the asthmatic patients CC was the prevailing genotype considering a significant difference when compared to control group (p=0.001) Table 2.


Figure 1: Gel image using QIAxcel System-genetic analyzer. Lanes 1-12 represent different samples. Lanes 1,6,7,8,12 representing the CC homozygote (two bands), lanes 4&10 representing the TT homozygote (one band) and products in lanes 2,3,5,9,11 representing the CT heterozygote (three bands). 15 base pair and 800 base pair bands are reference markers injected before the PCR products were injected.

  Control Asthma COPD
  No % No % No %
CC 11 18.3 26 52.0 15 37.5
CT 40 66.7 22 44.0 21 52.5
TT 9 15.0 2 4.0 4 10.0
χ2 (p)     14.976*(0.001) 4.642 (0.098)
Allele frequency            
C 51 51.0 48 52.2 36 59.0
T 49 49.0 44 47.8 25 41.0
χ2 (p)   0.026 (0.871) 0.980 (0.322)

Table 2: The genotype and allele frequencies of CD40 − 1C/T among the three studied groups.

CD40 -1C/T genotyping and its association with the risk for asthma and COPD: According to (Table 3A, 3B), The frequency of CC homozygous was significantly different from CT heterozygous in both asthma and COPD patients when compared with the control group. Being CC homozygous conferred a 4-fold increase in the risk of asthma (OR= 4.30, 95% CI: 1.79-10.32, p=0.001) and a 2.5-fold increase in the risk of COPD (OR = 2.60, 95% CI: 1.01-6.65, p = 0.044). The frequency of T allele was significantly different from C allele in asthmatic patients when compared to the control group (p=0.001), indicating that the risk of asthma was significantly lower (OR= 0.37) among individuals carrying T allele (95% CI: 0.21-0.66, p=0.001). However, in COPD patients no significant difference was observed between the C and T alleles.

  Control Asthma OR 95% CI P
  No. % No. %
CC 11 18.3 26 52.0 4.298 1.789 – 10.321 0.001
CT® 40 66.7 22 44.0      
TT 9 15.0 2 4.0 0.404 0.080 – 2.038 FEp = 0.320
CC + TT 20 23.3 28 56.0 2.545 1.173 – 5.523 0.017*
Allele frequency              
C® 62 51.7 74 74.0      
T 58 43.3 26 26.0 0.376 0.212 – 0.666 p# = 0.001*

Table 3A: The genotype and allele frequencies of CD40 − 1C/T between asthmatic patients and controls.

  Control COPD OR 95% CI P
  No. % No. %
CC 11 18.3 15 37.5 2.597 1.014 – 6.652 0.044*
CT® 40 66.7 21 52.5      
TT 9 15.0 4 10.0 0.847 0.233 – 3.078 FEp = 1.000
CC + TT 20 23.3 19 47.5 1.810 0.797 – 4.111 0.155
Allele frequency              
C® 62 51.7 51 63.8      
T 58 43.3 29 36.2 0.608 0.340 – 1.085 p# = 0.091

Table 3B: The genotype and allele frequencies of CD40 − 1C/T between COPD patients and controls.

Since there was a significant difference in the age, sex and smoking status between the asthmatics and controls (Table 1), therefore, we performed logistic regression analysis for the risk of asthma in the presence of these three covariates. Still, CD40 polymorphism exerted a significant effect on the risk of asthma where the frequency of CC of CD40 was significantly different from CT in the asthmatic patients (OR =0.029, 95% CI: 0.002-0.356, p = 0.006) Table 4A.

  Genotype Control Cases OR Sig.
No. % No. %
Asthma CC 11 18.3 26 52.0 0.029* (0.002 – 0.356) 0.006*
CT® 40 66.7 22 44.0    
TT 9 15.0 2 4.0 0.000 (-) 0.999

Table 4A: The comparison of genotype distributions between normal control subjects and subjects with asthma.

  Genotype Control Cases OR Sig.
No. % No. %
COPD CC 11 18.3 15 37.5 0.357* (0.130 – 0.975) 0.045*
CT® 40 66.7 21 52.5    
TT 9 15.0 4 10.0 0.307 (0.067 – 1.401) 0.127

Table 4B: The comparison of genotype distributions between normal control subjects and subjects with COPD.

While in COPD patients the smoking status was the only variable that differed significantly from the controls. (Table 1), therefore, we performed logistic regression analysis for the risk of COPD in the presence of this covariate. CD40 polymorphism also exerted a significant effect on the risk of COPD where the frequency of CC of CD40 was significantly different from CT in the COPD patients (OR =0.357, 95% CI: 0.130-0.975, p = 0.045) Table 4B.

No significant association was established between the genotype distributions and the smoking status nor the spirometric classification of severity in either asthmatic or COPD patients.

Effect of CD40-1C/T genotyping on CD40 protein expression: The level of CD40 expression on CD20+ B cells was significantly higher in asthmatic and COPD patients than in controls (p< 0.001, p=0.015 respectively) Figure 2, Figure 3.


Figure 2: CD40 expression measured by flow cytometry:
A: Dot plot of forward scatter against side scatter showing gated lymphocytes.
B: Dot plot of CD20 stained by (PE) and CD40 stained by (FITC) showing CD20+&CD40+cells in a studied case.


Figure 3: A comparison of CD 40 expression level among the three studied groups.
A: The data are presented as the ratio of the CD40 fluorescence intensity to that of CD20+cells. *:Statistical significance at p = 0.05

When CD40 expression levels were compared according to genotypic distribution, a significant difference was observed in the asthmatic as well as the COPD group in comparison to the control group (p=0.024 and p=0.008 respectively). Our data showed a greater amount of CD40 protein in the presence of C polymorphism as it was more expressed with the CT followed by the CC genotypes and the least expression was with the TT genotype (Table 5).

    Genotype χ2 (p)
    CC (n = 26) CT (n = 22) TT (n = 2)
Control CD 40        
Range 4.0 – 11.0 3.0 – 10.0 4.0 – 8.0 3.292 (0.193)
Mean ± SD 6.91 ± 2.30 5.83 ± 2.0 5.44 ± 1.33
Median 6.0 5.0 5.0
p1   0.102 0.112  
p2     0.713  
Asthma CD 40        
Range 2.0 – 18.0 2.0 – 18.0 5.0 – 5.0 7.421* (0.024)
Mean ± SD 9.35 ± 4.54 11.45 ± 4.08 5.0 ± 0.0
Median 9.0 12.0 5.0
p1   0.038 0.087  
p2     0.065  
COPD CD 40        
Range 3.0 – 15.0 2.0 – 17.0 3.0 – 4.0 9.652* (0.008)
Mean ± SD 7.67 ± 3.90 9.81 ± 4.41 3.50 ± 0.58
Median 7.0 10.0 3.50
p1   0.126 0.015*  
p2     0.003*  

Table 5: The comparison of CD40 expression levels on CD20+ B cells possessing different genotypes for -1C > T SNPs of CD40 in each of the studied groups

The total serum IgE and its relation to CD40-1C/T genotyping and CD40 protein expression in asthmatics: The total serum IgE was significantly higher in the asthmatic patients than the control group (p<0.001). The associations of CD40 genotypes and the total serum IgE levels was not statistically significant neither in subjects with asthma nor the control subjects when using the multivariate general linear model type III controlling for age, sex, and smoking status as covariates. Also there was no significant correlation between the CD40 expression level and total serum IgE in both groups.


CD40 is found on a variety of inflammatory cells and is known to influence the inflammatory state and play a role in airway inflammatory responses [3,13]. It is important for antibody class switching and is involved in the amplification and regulation of inflammatory immune responses, including regulation of T-cell-dependent B-cell responses and maturation of dendritic cells (DCs) [13] A C/T in the 5′-untranslated region of CD40 located at the − 1 position within the Kozak sequence (rs1883832) has been associated with CD40 protein expression [14,15].

These biological observations prompted us to evaluate the effect of CD40 -1C/T polymorphism on the risk for asthma and COPD. We found that this polymorphism was significantly different in the asthmatic group from the control group, independent of other confounding factors (age, sex and smoking status). Our study has demonstrated that CD40 − 1C/T polymorphism significantly contribute to the susceptibility to asthma in the Egyptian population, to our knowledge, this is a novel finding. Contrary to our results Park et al who studied asthmatic patients found that CD40 -1C/T polymorphism had no effect on the development of asthma in the Korean population
[ 16].This could be attributed to the different studied populations.

COPD is a complex disease influenced by genetic and environmental factors. Previous studies have shown that cigarette smoking is the major environmental risk factor. However, only about a minority (10–20%) of smokers develops the clinically significant disease, [17] also there are a considerable number of people who develop COPD without having smoked cigarettes [18] indicating that susceptibility to COPD may be influenced by genetic factors. We found that (independent of the presence of the main risk factor for COPD which is the smoking status) the CD40-1C/T polymorphism significantly contributes to the susceptibility to COPD, providing further support to the influence of genetic factors. Furthermore, Liu et al  who studied COPD patients got the same finding in the Chinese population, however, the odds ratio for CD40 -1C/T polymorphism was relatively lower (1.777) ,and thus they suggested that other genes may contribute to the genetic susceptibility to COPD [19].


CD40 -1C/T polymorphism affects the translational efficiency of CD40 protein giving rise to inflammatory responses in the airways and lung parenchyma [20]. This fact supports our findings in asthmatic and COPD patients where the CD40 expression was significantly higher than controls. The ribosome is able to initiate translation more efficiently in the presence of the C polymorphism and results in the formation of greater amounts of CD40 protein, [6] which may imply that the − 1C/T polymorphism is linked to inflammation in addition to the initiation and development of asthma and COPD. As our data have shown a greater amount of CD40 protein in the presence of C polymorphism as it was more expressed with the CT followed by the CC genotypes and the least expression was with the TT genotype, which seems to have a protective effect on the protein expression. Indeed, Jacobson et al have shown that the T allele caused a significantly reduced CD40 expression in B cells, which further supports our results [6].

Previous studies have shown that CD40 and CD40L interactions have been implicated in lung disorders, [21] giving rise to a decrease in FEV1 and FEV1/FVC. Interestingly, we did not find any association between the − 1C/T polymorphisms and the spirometric classification of severity of asthmatic as well as COPD patients, suggesting that CD40 sequence variances have no effect on lung function decline in either of them. Hence, CD40 − 1C/T polymorphism may act only as a potentiating factor in asthma and COPD, in concert with polymorphisms of other genes, as well as other environmental factors [19].

Despite the significant upregulation of total serum IgE in asthmatics it was not significantly associated with CD40 genotyping or the protein expression. Despite the fact that many studies have provided evidence that CD40 plays a role in the regulation of IgE [22]. However, our results did not show a direct link between the total serum IgE and the CD40-1C/T polymorphism. In addition, Park et al found that two CD40 SNPs, –1C/T and –580G/A, are associated with total IgE levels in individuals with asthma [16]. Further studies are needed to explore the gene–gene and gene–environment interactions involved in the development of asthma and in IgE regulation.

In conclusion, our study has demonstrated that CD40 − 1C/T polymorphism significantly contribute to the susceptibility to asthma and COPD in the Egyptian population supporting the stipulation that asthma and COPD can have common genetic risk factors. However, carrying T allele showed a significantly lower risk for asthma but not COPD. In addition, CD40 genotypes did not seem to be associated with the smoking status or clinical severity of both. Hence, CD40 − 1C/T polymorphism may act only as a potentiating factor in asthma and COPD, in concert with polymorphisms of other genes, as well as other environmental factors.

The genetic predisposition to certain pathways may further help to define the development of either asthma or COPD. In the end this may lead to stratification of patients by their genetic make-up and open new therapeutic prospects.


Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Relevant Topics

Article Usage

  • Total views: 13034
  • [From(publication date):
    December-2011 - Sep 21, 2019]
  • Breakdown by view type
  • HTML page views : 9152
  • PDF downloads : 3882