The functions of miRNAs can provide valuable diagnostic measurements as well as prognostic indicators for several diseases including types of cancer 5. A healthy liver consists of different cells producing a variety of miRNA expression profiles according to complex intrinsic and extrinsic signals. Under pathogenic conditions, these cells can also alter the profile of miRNAs 23. Commonly, HCC is accompanied by loss of liver and the remaining treatment would be transplantation. Although many studied have been accomplished in order to find early detecting markers on the basis of miRNAs, still no effective biomarkers have been introduced in this regard (6 of 2019, Jin). By the way, as HCC renders to transplantation, the maintenance of the new liver and prevention from rejection is the priority of these patients. In the present research, it is tried to evaluate the alterations in miRNA expression profile of LT patients with HCC in order to introduce probable miRNA biomarker in early detection of rejection. In this sense, LNA-array probe was used to identify the expression profile of the miRNAs and the results of LNA-array was tested in other samples of liver transplanted patients with the same condition of underlying disease. Although it has been suggested that some of the circulating blood miRNAs have the potential to be used as miRNA-based blood biomarkers in cancer detection 24,25, it is noteworthy to investigate the possibility of introducing more accurate miRNAs in detecting rejection in the biopsy samples that are taken commonly from patients. In a study done by Miyaaki et al, they used the same techniques as we did, for studying the miRNA profile of liver transplanted patients with HCV/HIV co-infection and confirmed that miR-101b and miR-149 are significantly decreased in co-infected group of patients comparing to HCV-infected ones 26.
The findings revealed that miR-4449, miR-4511, miR-3158-5p, and miR-4633-5p had been up-regulated and miR-4284, miR-122-3p, miR-194-5p, and miR-548as-3p had been down-regulated in patients with LT rejection once compared with non-rejected ones. The results from LT patients were also in accordance with the LNA-array probe results. With reference to the results collected from the PCA plot, there was a significant difference between miRNA expression in the rejected LT patients and the non-rejected ones.
The miR-122-3p is a liver-specific miRNA, highly expressed, comprised of more than 70% of all cloned liver miRNAs, unlike other tissues 1,27. Moreover, this miRNA plays a role in the liver metabolism, for instance, many genes interfering in lipid metabolism regulation can be down-regulated when using an antisense strategy to knock it down 28. The sequence and the expression pattern of the miR-122-3p in the liver is also highly conserved 29. In a study, HCC patients had shown a reduction in miR-122-3p expression 30. MiR-122-3p, miR-26a, and miR-195 have been also detected as tumor suppressors in the liver 31. The fact that miR-122-3p reduces in HCC patients was accordingly certified in the present study. Other studies on liver transplant recipients with recurrent hepatitis C and acute cellular rejection demonstrated the importance of miR-122 and miR-194 in liver transplant rejection either 24,32. Furthermore, the correlation between miR122-3p and -194-5p has been a subject of research studies. Therefore, this miRNA was analyzed between HCC non-rejected and rejected patients, indicating that the given miRNA could significantly (p=0.041) decrease in the rejected LT cases.
The miR-194-5p expression had been previously detected, reporting that such an miRNA could play a role in activation of stellate cells during liver fibrogenesis 33,34. Another study had further profiled the expression of miR-194-5p as a marker of liver HCC metastasis by showing the overexpression of miR-194 in cancerous liver cell lines to down-regulate N-cadherin expression and to suppress migration, invasion, and metastasis 31. An increase in the expression of miR-194-5p in HCV-infected hepatoma cells has been also well documented although the miR-194-5p expression level had significantly reduced in HCC patients 35. In addition, Farid et al. had demonstrated that the expression levels of miR-122 could significantly correlate with miR-194 among HCC patients (r=0.322, p=0.007). They had also reported no significant differences between non-rejected and rejected groups of patients against the expression of miR-194-5p (p=0.134). However, the difference in miR-122 expression level in their study was significant (p=0.001) 36. In addition, a research in 2015 had found that miR-194-5p had been down-regulated in the tissues involved in HCC and they had reported that it was able to reduce cell viability and proliferation by inducing G1 arrest and apoptosis in HCC cells. Moreover, they believed that miR-194-5p, inhibiting MAP4K4, was a potential diagnostic and therapeutic marker for HCC 37.
Some miRNAs might be originated from repetitive elements like transposable elements (TEs) 38. Hsa-mir-548as-3p is also a member of a large human gene family, derived from transposable elements named Made1 39. Approximately 69 members are found in almost all human chromosomes, especially chromosomes 6, 8, and X (30.43%). The predicted targets for hsa-miR-548as-3p are also functional molecules in many biological processes like MAPK signaling, cell cycle, p53 signaling pathway, colorectal cancer (CRC), non-small cell lung cancer, B cell receptor signaling pathway, transforming growth factor β (TGF-β) signaling pathway, renal cell carcinoma, and many others 40. TGF-β signaling pathway is also important in post-transplant inflammation. Besides, it is considered as one of the main mediators and inducers of fibrosis. Therefore, miRNA down-regulation during the rejection period renders to overexpressed SMAD4, enhancing the signaling pathway of TGF-β 41. In addition, it has been reported that Tg737 is a target gene for miR-548as-3p, whose down-regulation facilitates HCC cell proliferation in vivo and in vitro, followed by an increase in colony formation 42. The present study showed that the given miRNA had significantly (p=0.041) reduced in rejected LT patients with HCC, accounting for its critical role in rejection.
The miRNA-4248 is located on chromosome 7, but there is not much data published about it. One study had used cross-mapping to detect miRNAs and had reported hsa-miR-4284 in cross-mapping events 43. In addition, another research team had reported that such miRNAs could be the most down-regulated ones in the tissue samples of ulcerative colitis patients, and they had further claimed that miR-4284 level correlated with the disease activity by regulating CXCL5 mRNA expression 44. Another research had found that miR-4284 and -4484 could be considered as diagnostic biomarkers for diffuse large B-cell lymphoma (DLBL) disease 45. The down-regulation of this miRNA was further detected in LT patients with HCC in the present study. Furthermore, the sensitivity and the specificity of analyses of the four down-regulated miRNAs showed that miR194-5p and miR548as-3p might have enough qualities (Table 3) for being biomarker candidates for rejection especially among HCC patients.
In this research, a profile of miRNA was demonstrated in which miR-194-5p, -548as-3p, -122-3p, and -4284 were down-regulated and miR-3158-5p, -4449, -4511 and -4633-5p were upregulated in HCC patients with rejected LT. MiR-3158-5p is also located on chromosome 10 and has more than 500 predicted targets. The bioinformatic studies here showed that TGF-β2 gene could be observed among its upper 95% target scores (see Additional File 1, Table A1; www.miRDB.org). TGF-β also plays an important role in preparing a suitable microenvironment for tumor cell growth in liver diseases. This factor causes a signaling pathway that can promote progression of HCC in two ways; first, this factor has an intrinsic activity as an autocrine or paracrine growth factor, and second, it has extrinsic activity inducing changes in microenvironments like variations in cancer-associated fibroblasts, regulatory T cells (Treg cells), and inflammatory mediators 46. The other up-regulated miRNA is miR-4449, located on chromosome 4 with 22 predicted targets in miRDB (see Additional File 1, Table A2). Among the predicted targets, the cyclin-dependent kinase 5 regulatory subunit 2, p39 (CDK5R2) is detected as the most highly scored one. Previously, P39 had been introduced as a potential clinical prognostic marker for HCC 47. Also, it had been detected that miR-4449 was a potential blood-based marker in multiple myeloma 48. The present study suggested that these two mentioned miRNAs (i.e., miR-3158-5p and miR-4449) had significantly (p=0.014 and 0.012 respectively) up-regulated in HCC patients with LT rejection. Even in analyzing sensitivity and specificity, these two miRNAs had AUC greater than 0.8 (refer to Table 4 for further information); therefore, they were proposed as a biomarker candidate for rejection in HCC patients.
The third up-regulated miRNA in HCC patients with LT rejection was miR-4633-5p, located on chromosome 5 with more than 100 predicted targets in miRDB (see Additional File 1, Table A3). In 2015, a research group had further noted that miR-4633-5p could be significantly expressed in human epidermal growth factor receptor 2 (HER2)-positive breast carcinomas among 85 tested miRNAs versus normal people 49. The last up-regulated studied miRNA in LT rejected patients with HCC is miR-4511, located on chromosome 15 with more than 400 predicted targets in miRDB (see Additional File 1, Table A4). There are also few studies on the role of miRNA in LT and further studies are warranted to detect the exact role of this miR and its targets. The results of the present study showed the up-regulation of these two miRNAs in HCC patients with rejected LT.
It is of note that more than 97% of liver biopsies are hepatocytes and biliary cells, and the rest are Kupffer cells and endothelial ones. In LT rejected biopsies, some macrophages, plasma cells, and lymphocytes are also added to their population 50. Given the fact that the selected miRNAs are expressed in the non-rejected group of patients and the majority of cells in biopsies are hepatocytes, it can be concluded that alterations in the expression profile of the selected miRs might be due to the rejection process in hepatocytes.
Putting together, noticing the fact that there is un urgent need for detecting biomarkers in diagnosis and prognosis of HCC, the findings of this study might be the first step in this pathway and seems to be preliminary keys for further confirmation of their regulatory role in rejection in LT patients with HCC. For instance, among down-regulated miRNAs, miR194-5p and miR548as-3p, both showed significant difference between rejected and nonrejected samples and the ROC curve analysis of these two miRNAs showed significant high specificity and sensitivity. Also, among up-regulated miRNAs in this study miR-4449 and miR-3158-5p can be considered to have important roles in causing rejection in such patients. Besides the ROC curve analysis of these two microRNAs showed significant high specificity and sensitivity. Further studies on detecting the important targets of these miRNAs, obviously would be a large finding in this field.
Although finding miRNAs as biomarkers might be very helpful for more accurate prognosis and diagnosis of HCC, some limitations interfere in this type of researches like ours. One of them is the limitation that exist for samples size. As HCC is not a very common disease, collecting tissue samples from such patients takes time, and the other limitation is lack of follow ups due to the risky and invasiveness nature of biopsy procedure. Finally, according to mentioned points, it is not possible to take tissue biopsy samples from normal population. Due to tissue specificity of miRNAs, it would be helpful for a better comparison.