Diagnostic value of serum miR-145 and miR-185 as targeting of the APRIL oncogene in the B-cell chronic lymphocytic leukemia

Background: Chronic lymphocytic leukemia (CLL) is one of the most common hematologic malignancy in adults worldwide. This cancer has a poor prognosis at different stages. So, the identication of new biomarkers is important for early diagnosis of B-CLL. Considering the oncogenic role of APRIL molecule in this leukemia as well as the regulatory role of microRNAs (miRNAs) in different signaling pathways, the present study evaluated the miRNAs targeting APRIL gene in B-CLL. Methods: The miRNAs were predicted and selected using bioinformatics algorithms. A total of 80 plasma samples (40 samples of healthy individuals and 40 samples of B-CLL patients) were subjected to RNA extraction and synthesis of cDNA. The expressions levels of predicted miRNAs and APRIL gene in plasma of B-CLL patients and healthy individuals were assessed by Real time PCR analysis. ROC analysis was performed to investigate the role predicted miRNAs as novel biomarkers in diagnosis of B-CLL. Results: The results of the prediction showed that miR-145-5p and miR-185-5p target the APRIL gene. The expression level of APRIL gene was strikingly higher in plasma of B-CLL patients than in the healthy individuals (102, P= 0.001). On the other hand, expression levels of miR-145-5p and miR-185-5p were strikingly lower in B-CLL patients than in the healthy individuals (0.07, P= 0.001) (0.29, P= 0.001). Also, ROC curve analyses demonstrated that miR-145-5p and miR-185-5p are specic and sensitive and may serve as new biomarkers for the early detection of B-CLL. Conclusions: These data suggest that the study miRNAs may have a role in B-CLL development and progression. Moreover, miR-145-5p and miR-185-5p can be served as a novel and potential biomarker in the diagnosis of B-CLL. miRNAs targeting studies. In the present study, we predicted miRNAs targeting APRIL gene using bioinformatics software, and determined the expression levels of these miRNAs and the APRIL gene using RT-qPCR in the plasma samples of B-CLL patients and healthy individuals. ROC curve analyses were also performed to investigate the possibility of using plasma levels of predicted miRNAs as new and potential biomarkers for detecting B-CLL.


Background
Chronic lymphocytic leukemia (CLL) is one of the most common hematologic malignancy in adults worldwide, that makes up approximately 30% of all blood cancers (1). In this leukemia, mature and monoclonal B or T lymphocyte cells with speci c surface markers such as CD 5,19 and 23 accumulate in the peripheral blood (PB), lymphoid organs, bone marrow (BM) and spleen (2). The diagnosis of CLL is di cult due to the lack of speci c sings at different stages (3). So, the identi cation of suitable biomarkers for better diagnosis and appropriate treatment is important. At present, CLL is often diagnosed using a variety of methods such as complete blood count (CBC) and differential white blood cell count, morphological assessment of blood smear and immunophenotyping of PB (4)(5)(6). Other biomarkers including Zeta-chain-associated protein kinase 70 (ZAP-70), chromosomes abnormalities, serum thymidine kinase, β2-microglobulin and mutations in the immunoglobulin heavy chain variable gene (IGHV) and TP53 important in prognosis of CLL (7,8). Besides of these markers, microRNAs (miRNAs) with properties such as tissue speci city, rapid release rate and plasma stability have been identi ed as non-invasive biomarkers in diagnosis and treatment of CLL (9)(10)(11). These molecules regulate gene expression through target mRNA degradation or inhibition in translation level (12). They Also play a critical role in various processes of physiological and pathological such as cell growth and differentiation, metabolism, apoptosis, angiogenesis and tumorigenesis (13). The structure and function of miRNAs suggest that the expression of many miRNAs in cancerous tissues changes abnormally compared to normal tissue (14). Therefore, altering the expression pro le of miRNAs can be used for the detection of a wide range of diseases (15). Further, miRNAs are associated with signaling pathways that frequently change in different cancers such as CLL (16). Among signaling pathways involved in CLL pathogenesis, NF-kB signaling pathway has an essential role in disease development (17). In alternative (non-canonical) NF-kB signaling pathway, APRIL (a proliferation-inducing ligand) as one of the members of the tumor necrosis factor (TNF) family bind to tumor necrosis factor receptor (TNFR) such as TACI (transmembrane activator and calcium modulator and cyclophilin ligand interactor) and BCMA (B-cell maturation antigen) and causes proliferation, survival and protection CLL cells from apoptosis (18,19) (Fig. 1). APRIL is a soluble molecule secreted from the Golgi apparatus (20). This molecule not detectable in normal tissues but is overexpressed in tumor tissues. The various studies have shown that the level of APRIL in the serum of CLL patients are signi cantly higher than in normal individuals (21,22).
Subsequently, considering the role of APRIL molecule in the pathogenesis of CLL and also the regulatory role of miRNAs in many signaling pathways, the aim of present study was miRNAs targeting APRIL mRNAs evaluation in B-CLL.

Patients and Clinical samples
In total, 80 plasma samples were collected from Valiasr Hospital (Arak, Iran) and stored at -70˚C before analysis. The collected samples included 40 samples of healthy individuals as control group and 40 samples of patients. All patients diagnosed as B-CLL by FAB criteria with an expert hemato oncologist. In diagnosed of B-CLL patients, European Group for the Immunological Characterization of Leukemias (EGIL) suggested the ow cytometry marker to diagnosed positive for CD5, CD19, CD20 and CD23, the level of surface CD20, and CD79 are low in compare of normal B-cell also, B-cells are negative for CD13, CD117 and other myeloid linage markers.
Also, the B-CLL group did not receive any conventional treatment (surgery, chemotherapy and radiotherapy). In both groups, samples were taken from 20 males and 20 females, with a median age of 53 years (range, 30 to 67 years) in the group of patients and 34 years (range, 26-68 years) in Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js the group of healthy individual. All patients and healthy volunteers provided written informed consent. The characteristics of the patients under study and the control group are shown in Table 1. . Finally, two miRNAs as targeting for APRIL gene were selected based on higher scores and better binding and the sequences of these miRNAs were obtained using the miRBase () database.

Primers Design
In this study, three types of primers including reverse transcription-speci c stem-loop primers for the synthesis of cDNA from miRNAs, random hexamer primers for the synthesis of cDNA from mRNA and speci c primers for evaluation of the expression of miRNAs and APRIL gene were used. These primers were designed by the GeneRunner and AlleleID7 softwares and the speci city of primers designed was determined by NCBI BLASTn tool. Also, the relative expression of predicted miRNAs and APRIL gene were normalized against miR-103 and GAPDH of expression, respectively.
The list of used primers and their sequences illustrates in Table 2.

Results
MiR-145-5p and miR-185-5p as miRNAs targeting APRIL gene The results obtained from different bioinformatics databases including miRanda, TargetScan, miRwalk, DIANA, miRTarBase, miRDB, miRdSNP, miRecords, miRGate and miRGator demonstrated that four miRNAs were generally capable of binding to APRIL mRNA as shown in Table 3. Although different mathematical algorithms and scoring method applied in these websites, we made a score table base on highest repeat and best Loading [MathJax]/jax/output/CommonHTML/fonts/TeX/fontdata.js complementarities with the target gene. Therefore, only miR-145-5p and miR-185-5p had the highest repeat, score and best complementarities with the target gene and these two miRNAs were chosen for the subsequent experimental study. The probability of targeting 3`-UTR part of the APRIL transcript by miR-145-5p and miR-185-5p seed reigns was shown in Fig. 2.    Table 4.

Discussion
CLL is the most prevalent leukemia in adults which remains a public health problem in the world (24). This cancer has poor prognosis and diagnosis at different stages (25). In the past, it was di cult to diagnose B-CLL patients, but today it is partially detectable using symptoms and analysis of prognostic laboratory biomarkers (26). Therefore, the identi cation of new biomarkers with high sensitivity and speci city is crucial for early diagnosis of B-CLL. Several reports have shown that changes in miRNAs expression are related to the development and progression of different cancers in human (27). Additionally, miRNAs are stable in serum and plasma, which might be due to their protection by exosomes, microvesicles and Argonaute 2 (28). Therefore, these molecules as non-invasive biomarkers can be useful for detection of different cancers including B-CLL (29,30).
On the other hand, the APRIL is known as an oncogene that is rarely expressed in normal tissues but its high levels are detectable in tumor cells in vivo and in vitro and cancers such as thyroid and colon (31). Additionally, APRIL is a soluble molecule secreted from the Golgi apparatus (20).
Some studies have demonstrated that APRIL, as an oncogene, upregulated in non-canonical NF-kB signaling pathway and that dysregulation of this signaling pathway contributes to development and evolution of B-CLL (32). Therefore, considering the role of APRIL in the pathogenesis of B-CLL, this molecule seems to be a suitable choice for miRNAs targeting studies. In the present study, we predicted miRNAs targeting APRIL gene using bioinformatics software, and determined the expression levels of these miRNAs and the APRIL gene using RT-qPCR in the plasma samples of B-CLL patients and healthy individuals. ROC curve analyses were also performed to investigate the possibility of using plasma levels of predicted miRNAs as new and potential biomarkers for detecting B-CLL.
The bioinformatics predictions presented here indicated that four miRNAs including miR-145-5p, miR-185-5p, miR-6132 and miR-4644 could target the transcript of the APRIL gene. However, miR-145-5p and miR-185-5p had higher scores and were chosen for experimental study.
The results of RT-qPCR in the current study showed that the plasma levels of these two miRNAs were signi cantly lower in the B-CLL patients than in the healthy individuals. The expression of miR-145-5p and miR-185-5p were downregulated by -14.28 fold (P = 0.001) and − 3.44 fold (P = 0.001), respectively. On the other hand, the plasma levels of APRIL were upregulated by 102 fold (P = 0.001) in the B-CLL patients than in the healthy individuals. The decreased expression of miR-185 has been reported in the several solid tumors (33). Zanette et al., showed that ve miRNAs including miR-135b, miR199s, miR-142-5p, miR-181c and miR-185 had the lowest expression levels in CLL (29). The role of miR-185 has been investigated in cancers including lung cancer (34), osteosarcoma (35) and breast cancer (36) by targeting SOX9, HK2, DNMT1 and E2F6, respectively. On the other hand, miR-145 was reported as a tumor suppressor miRNA because its expression is reduced in various tumor, including colon, breast, ovarian, CLL and burkitt lymphoma (37,38). These nding suggest that miR-145-5p and miR-185-5p act as a tumor suppressor in the B-CLL by direct-targeting of APRIL mRNA. Therefore, B-CLL progression may be prevented by the expression of miR-145-5p and miR-185-5p in cancer cells through decreases in the expression of APRIL gene. The accuracy of miR-145-5p and miR-185-5p levels to discriminate between patients with B-CLL and healthy individuals using ROC curve analyses revealed that these two miRNAs are speci c and sensitive for the detection of B-CLL but miR-145-5p presented the greatest AUC, sensitivity, and speci city as 90%, 95%, and 0.952, respectively, at a cutoff of 0.034. Therefore, differences in the expression of miR-145-5p could distinguish healthy individuals from patients with B-CLL. Akao et al., also con rmed that miR-145-5p can be used as biomarker to differentiate normal cells from malignant B cells (38). Therefore, according to the ndings of present study, it can be concluded that serum miR-145-5p and miR-185-5p levels could be applied as potential and e cient biomarkers in B-CLL.

Conclusion
The results of the present study demonstrated a negative correlation between downregulation of miR-145-5p and miR-185-5p with upregulation of APRIL in B-CLL patients. This correlation may be due to the targeting of the APRIL mRNA (as an oncogene) by miR-145-5p and miR-185-5p (as tumor suppressor miRNAs). In conclusion, these miRNAs may be serve as new and non-invasive biomarkers in the diagnosis of B-CLL. Therefore, this study might provide a new diagnosis approach for B-CLL but more investigations are needed for discovery of functional relationship between identi ed miRNAs and the APRIL gene.

Consent for publication
All patients and healthy volunteers provided written informed consent Availability of data and material Please contact author for data requests.

Competing interests
The authors declare that they have no con ict of interests.

Funding
No funding sources used in this study. M.M. provided critical reviews in order to promote the manuscript. All authors consent for publication in BMC Cancer journal.

Figure 1
Schematic representation of the NF-kB signaling pathway activated by APRIL ligand in the CLL cells.