The impact of variations in transcription of DICER and AGO2 on exacerbation of childhood B‐cell lineage acute lymphoblastic leukaemia

Summary

The expression of microRNA in eukaryotic cells is subject to tightly regulated processing. The altered expression of microRNAs in a number of cancers suggests their contribution to disease pathogenesis, where processing pathways may be involved in disease pathogenesis. In the present study, we evaluated changes in the profile of two main components of microRNA biogenesis, AGO2 and DICER, and assessed their correlation with disease progression in childhood acute lymphoblastic leukaemia (ALL). To achieve this aim, 25 patients afflicted with ALL were included in the study along with 25 healthy subjects as control. The expression level of AGO2 and DICER was evaluated by real‐time PCR. The results revealed an increase in the expression of DICER and a decrease in AGO2 in patients. The correlation between the alteration levels of these genes with pathologic events was also studied. This increase or decrease proved to be directly correlated with the progression of the disease particularly in L1 to L2. According to the obtained results, it can be deduced that dysregulation in transcription of DICER and AGO2, involved in the formation of mature microRNAs in cytoplasm of ALL cancer cells, is a part of the pathological molecular mechanism implicated in the exacerbation of this malignancy. Therefore, the genes involved in microRNAs biogenesis that have been studied here could be considered as candidate prognostic markers especially in childhood ALL which will help towards a better understanding of the molecular basis of ALL.

1. INTRODUCTION

Acute lymphoblastic leukaemia (ALL) is a haematopoietic malignancy, which often afflicts children.1 The survival rate for children aged two to five years is approximately 90%. However, the chance of post‐therapeutic survival among those below one year of age and the patients above 10 years old is far lower.1, 2, 3 In ALL, the precursor B and T cells grow in an uncontrolled manner and increase in number. They reduce the number of other natural blood cells and eventually cause different infections in the affected individual.4 Various factors especially environmental and genetic factors can cause ALL. However, these factors and pathological molecular events involved in leukaemia are unknown or not well‐documented.

MicroRNAs are small non‐coding RNAs whose length is approximately 18 to 25 nucleotides, involved in the regulation of gene expression through degrading mRNA or inhibiting target mRNA translation. Over 50 000 copies of microRNA can be expressed in a cell, and approximately 1500 human microRNAs have been estimated to be capable of regulating 30%‐60% of all mRNAs. MicroRNAs play a vital role in differentiation, apoptosis and proliferation under normal body conditions. Their improper expression or regulation, however, can cause cancer. This would further highlight the role of miRNAs as an oncogene or a tumour suppressor.5 Recent studies have shown that microRNAs can basically play a role in the formation and expansion of cancer and consequently in the prognosis and treatment of the disease. For instance, changes in expression of miRNA‐99a and miRNA‐100 have been associated with ALL patients’ survival, and in vitro restoration causes suppression of cell proliferation via regulation of signalling pathways.6 Therefore, it can be hypothesized that changes in the regulation of the genes involved in the biogenetic pathway of microRNAs can play a key role in cancer and can be considered a candidate marker for progression or development of cancers.

DICER plays a key role in the biogenesis pathway of microRNAs. It is an enzyme from the RNase‐III family acting on double‐stranded RNAs. After driving of Drosha and DGCR8 on long pri‐microRNA, DICER creates a cut in the precursor of microRNAs and finally leads to maturity and function of microRNAs. Elevated levels of DICER have been reported as a poor prognostic factor in patients with prostate cancer7 and adenocarcinoma as well as melanoma.8 On the other hand, diminished expression of DICER has been reported in lung 9 and ovarian10 cancers.

Recent investigation has revealed that DICER is involved in the immune system. Rats with DICER knock‐down showed defects in T lymphocyte differentiation and regulation of T cell development. It was also observed that changes in DICER expression and finally in microRNAs could lead to disorders in maturation of B cells and selection of thymic T cells as well as in the sensitivity of T cells to antigens.11 Therefore, any significant alteration in the expression of DICER and other enzymes involved in microRNA biogenesis could contribute to pathogenesis of certain cancers such as leukaemia.

MicroRNAs need to join RNA‐induced silencing complexes (RISCs) to exert their function. AGO2 enzyme belongs to the Argonaut family (Ago) as the key element of RISC. The main function of AGO2 is its central role in RNA silencing processes, as an essential component of RISC. It binds to microRNAs to target mRNA and finally silence the gene. Further, in human cells, the microRNA‐Ago complex can enter the nucleus and bind to the promoter sequence, heterochromatinizing the structure of the target gene, which ultimately leads to silencing the transcription of the target gene.12

As mentioned above, dysregulation of microRNA levels could potentially be one of the causes of pathogenesis of multifactorial diseases such as cancer. They can also be used in the future as a diagnostic or prognostic biomarker or as a target to treat cancers. Therefore, the factors involved in the biogenesis pathway of these small non‐coding RNAs particularly involved in the formation of mature microRNAs would be of great significance. The present research investigated the changes in the transcription level of DICER and AGO2, two effective proteins involved in the cytoplasmic part of biogenesis of microRNAs, in patients with ALL.

2. METHODS

2.1. Sample collection

Blood samples from patients afflicted with ALL were obtained from the Pathology Department of Shahid Mohammadi Hospital and Pediatric Hospital located in Bandar Abbas with an informed consent signed by the participants’ parents before enrolment.

2.2. Ethical approval

The present study was approved by the Ethics and Human Rights Committee of Hormozgan University of Medical Sciences (HUMS), Bandar Abbas, Iran (HUMS.REC.1394.84).

2.3. Eligibility criteria

Definitive diagnosis of the disease was made by pathologists and oncologists through the results of flow cytometry, immunohistochemical tests and blood cell count as well as peripheral blood smear test. The control subjects were selected according to their blood cell count and peripheral blood smear test, who were confirmed as healthy, and matched according to age, gender and even ethnicity. The present study had time and place limitations. The samples should have been collected in Hormozgan province within two years. Finally, 25 subjects (afflicted with ALL) were assigned to the patient group and 25 healthy ones to the control group. None of the included subjects received radiotherapy or chemotherapy. All patient were Philadelphia (Ph) negative. All patients treated in some way prior to the study were excluded. The disease diagnosis and classification was determined based on French‐American‐British (FAB) classification as ALL‐L1, L2 and L3 and WHO classification of tumours of haematopoietic and lymphoid tissues as ALL and subtype to pro‐B, common and pre‐B‐cell ALL.13 Patients were diagnosed according to immunophenotyping by panel of antibody in flow cytometry study, cytomorphology, cytochemistry and bone marrow biopsy immunohistochemistry. Accordingly, patients were also categorized to three groups ‐ good, intermediate and poor ‐ with respect to leucocyte count, bulky extramedullary disease, CD10 negativity, immature pro‐B phenotype,14 and favourable treatment prognosis. This is equivalent to very high risk, high risk and standard risk, respectively, according to age (10 months > age > 10 years, 10 months < age ≤ 10 years), first leucocyte count at diagnosis (WBC count < 50 000/μL or ≥50 000/μL), subtype of ALL, gender, spread to certain organ sites (cerebrospinal fluid; testis in males), response to induction therapy as indicated by the minimal residual disease (MRD) and relapsing. Table 1 shows the subjects’ information for this study.

Table 1.

Demographic data of ALL patients and healthy controls

Variable	ALL patient (n = 25)	Healthy control (n = 25)	P‐value
Mean age (±SD) (Min‐Max)	6.70 (±3.42) (2.0‐13.5)	6.78 (±3.38) (2.0‐13.5)	0.931
Age ≤ 7	16 (51.6%)	15 (60.0%)
Age > 7	9 (47.4%)	10 (40.0%)
Gender
Female	10 (40%)	10 (40%)	1.001
Male	15 (60%)	15 (60%)
Mean % Blast Cells (±SD) (Min‐Max)	80.05 (±8.05) (64‐95)	0
% Blast < 80	10 (40.0%)	0
% Blast ≥ 80	15 (60.0%)	0
ALL FAB sub‐classification
L1	13 (52.0%)	0
L2	12 (48.0%)	0
L3	0	0
ALL EGIL sub‐classification
B‐I: Pro B‐ALL	11 (44%)	0
B‐II: Common B‐Cell	0	0
B‐III: Pre B‐ALL	14 (66%)	0
B‐IV: Mature B‐ALL	0	0
Favourite in treat prognosis
Good	3 (12.0%)	0
Intermediate	17 (68.0%)	0
Poor	5 (20.0%)	0

Abbreviation: EGIL, European Group of Immunophenotyping Leukemia.

2.4. Quantitative Reverse Transcription‐Polymerase Chain Reaction (QRT‐PCR)

RNA was extracted from the whole blood using ZR whole Blood RNeasy mini kit (Cat. No. R1020, Zymo Research Co.), and the instruction in the kit protocol was followed. The quality and quantity of the extracted RNA were evaluated with agarose gel electrophoresis and NanoDrop (BioRad), respectively, and consequently, the amount of extracted mRNA was calculated in ηg/μL. The reverse transcription was run on equal amounts of mRNA of each sample in the base of the kit protocol using an RevertAid™ First Strand cDNA Synthesis Kit (Cat.No. K1632, Fermentase). β‐actin gene was selected as the housekeeping gene. Table 2 presents the sequence of the primers.

Table 2.

Primer sequences and real‐time PCR conditions

Gene	Sequence of primers (5′→3′)	Amplicon (bp)	Annealing temp. (°C)
DICER	F‐ TGGGTCCTTTCTTTGGACTG	245	62
	R‐ CTGGTTTGCAGAGTTGACCA
AGO‐2	F‐ AGGCTGCTCTAACCCTCTTG	114	60
	R‐ ACGCTGTTGCTGACACATC
β‐actin	F‐ GCCTTTGCCGATCCGC	90	59
	R‐ GCCGTAGCCGTTGTCG

The initial denaturation at 95°C for 30 seconds was followed by 40 cycles via denaturation at 95°C for 10 seconds, thereby annealing temperatures (Table 2) for 30 seconds and extending at 72°C for 30 seconds. The mean expression of the housekeeping gene β‐actin was used as a control to normalize the variability in the expression level of DICER and AGO2 genes. We also used a no template sample as the negative control. A standard curve was included in each run for assay validation. For standardization of real‐time PCR, the serial dilution of the PCR product of the genes was used. For real‐time PCR performance, SYBR^® Premix Ex Taq^™ II (Tli RNaseH Plus) kit (Takara Bio Inc.) was utilized coupled with the Rotor‐Gene^™ 6000 system (Corbett Research). Every sample was run in triplicates.

2.5. Statistical analysis

2^−∆∆CT method was used to evaluate gene expression. The statistical analysis was performed using GraphPad Prism‐5, SPSS‐21 and Microsoft Excel software, where P‐values less than 0.05 were considered statistically significant in our experiments. In addition, Kolmogorov‐Smirnov test was used to check the data distribution and their normality. Accordingly, proper parametric and non‐parametric tests were run to compare the data. Further, Student's t test, Mann‐Whitney U‐test and Wilcoxon two‐sample test were applied to detect significant differences in the DICER and AGO2 expression level.

3. RESULTS

In the present study 25 patients with lymphocytic leukaemia participated, while 25 healthy subjects were considered as the control group. The mean age of the healthy group was under 7 years and the same as the patient group. Regarding age, the patients were divided into two groups, ≤7 and ˃7 years old. A total of 40% of the patient group and healthy controls were female, and the rest (60%) were male. The calculated P‐value attested that our collected healthy controls were age‐ and sex‐matched with subjects (Table 1). A total of 40% of patients had a blast cells lesser than 80%, and 60% of them had a blast cells above or equal to 80%. According to FAB classification, ALL entails three levels, L1, L2 and L3 corresponding to the cytomorphology of blast stained with Wright‐Giemsa method, risk stratification obtained from factors such as age, gender, white cell count and cytogenetic features reported by pathologists and oncologists. In the present research, the highest level of the disease was L2 (afflicting 48% of patients). The rest were classified as L1. There was no ALL‐L3 identified during our study. Considering prognosis and response to treatment, 12% of the patients enjoyed a good prognosis, while the rest had a poor (20%) or average (68%) rate as outlined in the classification above (Table 1). Unfortunately, two child patients with a poor prognosis died.

DICER expression was explored among ALL patients and compared to the control. The results obtained from real‐time PCR revealed a significant increase in DICER level among patients by 1.68 times, in comparison with the healthy group (P ˂ 0.0001) (Figure 1A). There was also a positive correlation between DICER level increase and the disease stages (particularly transitioning from L1 to L2) (Table 3, Figure 2A). The DICER expression was shown to be higher, respectively, in L2 than L1 and in L1 than the healthy samples. It was also found that DICER expression significantly rose in both L1 and L2 patients as compared with the healthy subjects (P = 0.0003). On the other hand, increased gene expression in L2 as compared to L1 was not statistically significant (P = 0.138). Also, a significant difference was calculated in expression of DICER between patients in group of blast cells ≥ 80% and patients with a percentage of blast cells lesser than 80% (P = 0.0366) (the data not shown). Significant increase in mRNA expression of DICER was also observed in the higher risk group of both Pro B‐ALL and Pre B‐ALL group (Table 3, Figure 2C).

Figure 1.

A, DICER; B, AGO2 expression level in ALL patients in comparison with healthy controls (H.C)

Table 3.

Real‐time PCR results for DICER and AGO2 mRNA expression at L1 and L2 stages of ALL

Subtypes	Mean (±SD) (Min‐Max)	P‐value
For DICER expression in blood
Normal	1.07 (±0.399) (0.62‐1.82)
L1	1.53 (±0.306) (0.94‐2.11)	0.0042
L2	1.98 (±0.764) (1.31‐3.50)	0.0003
Pro B‐ALL & High Risk	2.41 (±0.339) (1.51‐3.41)	<0.0001
Pro B‐ALL & Std. Risk	1.83 (±0.127) (1.22‐2.45)	0.0072
Pre B‐ALL & High Risk	2.20 (±0.683) (0.94‐3.50)	0.0031
Pre B‐ALL & Std. Risk	1.46 (±0.223) (1.16‐1.75)	0.0369
For AGO2 expression in blood
Normal	1.17 (±0.501) (0.56‐1.95)
L1	0.61 (±0.202) (0.39‐0.99)	0.0025
L2	0.39 (±0.090) (0.23‐0.52)	<0.0001
Pro B‐ALL & High Risk	0.28 (±0.041) (0.25‐0.36)	<0.0001
Pro B‐ALL & Std. Risk	0.58 (±0.191) (0.39‐0.78)	0.0024
Pre B‐ALL & High Risk	0.42 (±0.068) (0.23‐0.62)	0.0083
Pre B‐ALL & Std. Risk	0.71 (±0.111) (0.43‐0.99)	0.0104

Figure 2.

The comparison of mRNA expression of DICER (A,C) and AGO2 (B,D) in patients with different subtypes in base of FAB (A,B) and EGIL (C,D) sub‐classification as compared to healthy controls (HC), Std, standard

Conversely, a significant decrease (P ˂ 0.0001) was observed in the expression of AGO2 for half of the ALL patients. The expression of this gene was reduced by 0.49 time in patients, as compared to the healthy control (Figure 1B). The AGO2 expression was also observed to be lower in L2 than L1 and lower in L1 than the control group (Table 3 and Figure 2B). Reduced AGO2 expression in the patient group as compared to the control was statistically significant (P ˂ 0.0001 for L2 and P = 0.0108 for L1). No significant difference, however, was detected in AGO2 expression between patients with the percentage of blast cells ≥ 80 and < 80 (P = 0.8026). Finally, it was found that patients’ age and gender did not significantly affect DICER and AGO2 expression levels in the present study. Interestingly, the level of AGO2 was significantly lower in high‐risk group of both Pro B‐ALL and Pre B‐ALL as compared to intermediate risk, which are referred to here as standard risk (Table 3, Figure 2D).

4. DISCUSSION

The role of microRNAs in regulation of gene expression, and the correlation between changes in their level and incidence of a disease has been well‐established. Concerning ALL, the expression of several microRNAs (eg miR‐181a, miR‐181b, miR‐17, miR‐18a, miR‐19a, miR19b‐1, miR‐20a, miR‐21, miR‐92‐1, miR‐142, miR‐223 and miR‐150) has been indicated to alter in some stages of ALL. It seems that the changes can be regarded as diagnostic and prognostic markers in malignancies.15, 16 AGO2 and DICER are the central components of the microRNA biogenesis and regulation pathway. Therefore, extensive studies have been evaluating the changes in the expression of these genes across different cancers and their role in the pathobiology of cancers. All these factors encouraged the present research to examine AGO2 and DICER expression in patients in comparison with a healthy group, since we hypothesized that any alteration in these miRNA biogenesis components could be the reason of changes in overall cell microRNAs.

DICER plays a crucial role in the microRNA processing network and has been studied in several cancers, particularly in solid tumours. DICER lies in the chromosome location of 14q32.13 and has been modified in some cancers. In cancers such as neuroblastoma and breast cancer, DICER gene has been revealed to be deregulated at transcription and splicing levels.17, 18 Nevertheless, DICER‐related molecular cancerous mechanisms have not been identified yet. With this concern in mind, using microarray, Chiosea et al9 found that the locus of DICER was eliminated in lung adenocarcinoma and pre‐cancerous lesions. In a number of cancers, including ovarian cancer, DICER has been observed to undergo single nucleotide mutations, which might have induced the cancer.10 DICER knock‐down human embryonic stem cells (hESCs) grew significantly slower than the control cells. The knock‐down of DICER attenuated cell division in hESCs via perturbation of some miRNAs such as miR‐195 and miR‐372, regulating two tumour suppressor genes, respectively: CDKN1A, which encodes p21 regulating the G1/S transition, and WEE1, which encodes a negative G2/M kinase modulator of the CycB/CDK complex. Therefore, it can be deduced that DICER can negatively control cell cycle modulators at two phases of the cell cycle to ensure appropriate replacement of the stem cell population.19

The conclusion of another research group was that silencing DICER inhibits cell proliferation and promotes apoptosis in leukaemia cell lines. Also, DICER was found to be upregulated by GATA1, a zinc‐finger transcription factor, in AML patients. These data suggest that DICER plays an important role in AML, and upregulation of DICER1 induced by GATA1 may provide a cue for understanding the role of DICER expression levels in multiple types of cancer.20 Furthermore, cyclin D1‐mediated migration and cellular proliferation are DICER dependent. Therefore, it seems that cyclin D1 via a cdk‐independent mechanism affects microRNA biogenesis through DICER. It was also observed that cyclin D1 and DICER preserve heterochromatic histone modification (Tri‐m‐H3K9).21

No previous well‐documented study was found in the literature regarding DICER profile in leukaemia, particularly in childhood ALL. Only one research study was found in blood cancers.22 Therefore, the present study is among the limited body of literature on DICER expression in haematopoietic cancers, especially concerning ALL in children. The present findings indicated elevated DICER expression across ALL patients as compared to their healthy peers. This finding was in agreement with other studies on mucoepidermoid carcinoma (MEC),23 colorectal cancer (CRC)24 and non‐small cell lung cancer (NSCLC).25 Miller et al26 stated that children with >25% L2 lymphoblasts had a significantly higher relapse rate and significantly poorer survival. We also observed that DICER expression at mRNA level in L2 increased from L1. It can be deduced that DICER plays a role in the formation and progression of ALL and its advancement to more aggressive forms. It seems that the expression of DICER in solid tumour samples is reduced in some cases such as ovarian,27 gallbladder28 and breast cancers.29 However, various expression profiles for DICER have been reported in different cancers. The reason can be attributed to the fact the modification of DICER level which can occur at both transcription and post‐transcription stages mechanisms, significantly influencing the DICER level.

AGO2 is a component of miRNA synthesis machine in the cytoplasm. It is rational to assume that any changes in its expression level can epigenetically induce changes in microRNAs balance in malignancies. The present findings revealed that AGO2 expression was significantly reduced in the patient group as compared to the healthy group. In the comparison of AGO2 expression in L1 and L2, it was found that a significant decrease in the expression of AGO2 coincided with progression of leukaemia from L1 to L2. This finding was similar to what Völler et al30 observed in their study about protein level of AGO2 among those afflicted with melanoma. A web‐based search indicated that no academic study has evaluated AGO2 expression in childhood ALL. Thus, the present research is pioneering in the regard.

At a molecular level, a review of the related literature shed no light on the actual causes of changes in AGO2 expression. Few studies have tried to reveal the mechanisms involved in AGO2 expression. Accordingly, several possible mechanisms are suggested. The possible reason for the reduced levels of AGO2 in ALL could be explained according to fast degradation by the ubiquitin degradation pathway.31 Cheng et al32 reported an increase in AGO2 expression in cells and samples of hepatocellular carcinoma. Through an immunoprecipitation of chromatin, it was indicated that AGO2 could directly bind to the focal adhesion kinase (FAK) promoter as a vital molecule related to tumour metastasis initiating its transcription.32 In another study AGO2 played a considerable role in regulating the FGF2 gene and stabilizing microRNAs. FGF2 serves its main functions in differentiation, proliferation, migration and angiogenesis. An increase in FGF2 expression easily influences the emergence, progression and metastasis of cancer.33 In addition, Shen et al34 observed increased levels of EGFR as a well‐recognized oncogene in human cancers, particularly in hypoxic conditions. FGFR is capable of phosphorylating AGO2 in Tyr393. It consequently reduces AGO2 activity towards DICER and inhibits the processing of microRNAs from pre‐miRNA to a mature miRNA. Then, under the particularly prevalent hypoxic conditions, seen especially in solid tumours, inhibited maturation of microRNAs through AGO2/EGFR at the post‐transcription level can occur this finally intensified the survival and metastasis of the tumour cells34, 35 through defects in maturation of microRNAs, acting as a post‐transcriptional regulating system. According to the results obtained by Shen et al34 reduction in AGO2 expression in ALL may have influenced microRNA processing and limited the formation of mature microRNAs, which in turn, added to the severity of ALL under increased EGFR and hypoxic conditions.34, 35 Therefore, the survival of cancerous cells was prolonged, leading to, progression and aggravation of the leukaemia. This hypothesis, however, needs to be further investigated in the future.

Overall, further studies on more patients from different ethnicities are suggested by this study. Further work is also required to better understand the interaction of these proteins with cell cycle control and other mechanisms such as apoptosis and proliferation.

Thus, in summary AGO2 expression at mRNA level was downregulated and significant upregulation of DICER mRNA expression was observed in ALL patients compared with healthy controls. It seems DICER may act like a cell cycle promoter factor as the studies on DICER knock‐down or silencing reported the attenuation of cell cycle and proliferation.19, 20 It was assumed that dysregulation in factors involved in the biogenesis of micro RNA may affect the level of microRNAs. As noted above this aberrant regulation in the biogenesis of microRNAs can contribute to disorders in other cellular pathways particularly in favour of cancer cells. It seems that dysregulation in the level of these components may play a role in the emergence and development of the disease via impairment in the cell cycle, control of tumorigenesis, and apoptosis. Since it was clear that greater variations in the transcription level of these factors involved in microRNAs biogenesis coincided with the progression and development of this cancer, that these factors can be considered as a candidate indicator for the progression of ALL or even as a prognostic marker.

CONFLICT OF INTEREST

All authors of this article have no conflict of interest.

Piroozian F, Bagheri Varkiyani H, Koolivand M, et al. The impact of variations in transcription of DICER and AGO2 on exacerbation of childhood B‐cell lineage acute lymphoblastic leukaemia. Int. J. Exp. Path. 2019;100:184–191. 10.1111/iep.12316