Identification of Recurrence-Related mRNAs and Noncoding RNAs in Hepatocellular Carcinoma Following Liver Transplantation

Background: This study aimed to determine whether altered mRNA, long non-coding RNA, and circular RNA expression is related to hepatocellular carcinoma recurrence after liver transplantation. Methods: Five recurrent and 5 non-recurrent primary hepatocellular carcinoma samples were used to perform RNA sequencing. Then, differentially expressed mRNAs, differentially expressed long non-coding RNAs, and differentially expressed circular RNAs between recurrent group and non-recurrent group were identified. In addition, differentially expressed long non-coding RNA/differentially expressed circular RNA–differentially expressed mRNA co-expression network, and competing endogenous RNA (differentially expressed circular RNA/differentially expressed long non-coding RNA–miRNA–differentially expressed mRNA) regulatory network were constructed. Finally, real-time quantitative polymerase chain reaction was performed for verification. Results: Five hundred forty-one differentially expressed mRNAs, 239 differentially expressed long non-coding RNAs, and 16 differentially expressed circular RNAs in the recurrent group were obtained. Gene set enrichment analysis indicated that these differentially expressed mRNAs may affect hepatocellular carcinoma recurrence through multiple pathways, such as E2F, epithelial–mesenchymal transition, G2M, and oxidative phosphorylation. Then, 993 differentially expressed long non-coding RNA–differentially expressed mRNA co-expression pairs and 28 differentially expressed circular RNA/differentially expressed mRNA co-expression pairs were obtained. The competing endogenous RNA network contained 10 circular RNA–miRNA pairs, 12 long non-coding RNA–miRNA pairs, and 36 miRNA–mRNA pairs. Real-time quantitative polymerase chain reaction results showed the same expression trend as RNA-seq results. Conclusion: Our results reveal key mRNAs, long non-coding RNAs, and circular RNAs associated with recurrence in hepatocellular carcinoma patients after liver transplantation and lay the foundation for understanding the molecular mechanism of hepatocellular carcinoma recurrence after liver transplantation.


INTRODUCTION
Surgical resection and liver transplantation (LT) are common treatments for hepatocellular carcinoma (HCC). Hepatocellular carcinoma has a higher recurrence rate. However, the precise mechanisms underlying the recurrence of HCC are still unknown. Therefore, identifying reliable and accurate predictive markers to screen out which subset of patients with HCC is vulnerable to develop recurrence is urgently needed.
Studies have found that long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) are involved in the development of HCC. MALAT1 could enhance hepatocarcinoma cell growth. 1 The cSMARCA5 inhibits the growth and migration of HCC cells and is a potential therapeutic target. 2 The circAKT3 is associated with HCC recurrence and mortality. 3 In the present study, we analyzed the expression of lncRNAs, circRNAs, and mRNAs in HCC recurrence after LT to obtain the differentially expressed (DE) mRNAs, DElncRNAs, and DEcircRNAs, aiming to identify early-phase biomarkers to predict HCC recurrence after LT. In addition, DElnc RNA/D Ecirc RNA-D EmRNA co-expression network and competing endogenous RNA (ceRNA, DEcir cRNA/ DElnc RNA-m iRNA-DEmRN A) regulatory network were constructed.
Through this work, we hope to provide some insights into the personalized treatment of HCC and improve the strategy of postoperative recurrence monitoring in HCC patients.

MATERIALS AND METHODS Subjects and Samples
Five recurrent and 5 non-recurrent primary HCC samples obtained from 10 patients who underwent LT were included in the study. All samples were collected after obtaining written informed consent from every participant. This study was approved by the ethics committee of the Third Hospital of Hebei Medical University and performed in accordance with the Declaration of Helsinki. Peripheral whole blood (2.5 mL) drawn from each subject was used for RNA extraction.

RNA Sequencing, Identification of Differentially Expressed mRNAs, Differentially Expressed Long Non-coding RNAs, and Differentially Expressed Circular RNAs
With PAXgene® RNA blood tubes, total RNA was extracted from samples. Based on the DNBSEQ platform (PE100 strategy), sequencing was performed. The Fastp was used to trim 5ʹ and 3ʹ segments of reads to remove bases with mass <20 and delete reads with N >10%. HISAT2 was applied to align the clean reads with the human reference genome (GRCh38). The expression of mRNAs and lncRNAs was normalized and outputted with StringTie. Then, the CIRIquant software was used to predict circRNAs. DEseq2 was used to identify DEmRNAs, DElncRNAs, and DEcircRNAs in recurrent HCC (reHCC) with |log 2 FoldChange| (|log 2 FC|) >1 and P < .05.

Differentially Expressed Circular RNA/Differentially Expressed Long Non-coding RNA-miRNA-Differentially Expressed mRNA (Competing Endogenous RNA) Network
The miRWalk was used to predict the target DEmRNAs of miRNAs. The NPInter v4.0 was used to predict the DElncRNA-miRNA interaction pairs. The DEcircRNA-miRNA interaction pairs were predicted with targetscan. Then, circRNA/lncRNA-miRNA pairs were combined with miRNA-mRNA pairs to construct the ceRNA regulatory network.

Real-time Quantitative Polymerase Chain Reaction
Four recurrent and 6 non-recurrent primary HCC samples obtained from 10 patients who underwent LT were enrolled in this study. Following the manufacturer's protocol, total RNA was isolated from blood samples with the RNAliquid reagent. The real-time quantitative polymerase chain reaction (RT-qPCR) was performed using SuperReal PreMix Plus (SYBR-Green; Tiangen, Beijing, China). Glyce ralde hyde-3-pho sphat e dehydrogenase (GAPDH) and actin beta (ACTB) were used as endogenous controls. Relative gene expression was analyzed by the 2 -ΔΔCT method.

DISCUSSION
Liver transplantation remains one of the most curative treatment options for HCC, with increasing mortality and morbidity of HCC. Hepatocellular carcinoma recurrence is the main underlying cause for the poor prognosis of HCC following LT. Herein, we identified the genes lncRNAs and circRNAs that are related to tumor recurrence in HCC after LT.  13 Teng et al 14 suggested that IDI2-AS1 was related to bone metastasis in breast cancer. These lncRNAs mentioned above were dysregulated in this analysis, indicating they may be involved in HCC recurrence after LT.
Genomic imprinting is one epigenetic phenomenon which is associated with many human diseases or syndromes as well as with various types of cancers. In individuals with a paternally imprinted gene, only the allele inherited from the mother is expressed and vice versa. The PPP1R9A gene is located in a cluster of imprinted genes on human chromosome 7q21, which was upregulated in hepatosplenic T-cell lymphoma. 15 Downregulation of PPP1R9A was correlated with more advanced cancer stages in squamous cell carcinoma of the head and neck. 16 The reduced expression of miR-22-3p in breast cancer was associated with tumor size, tumor node metastasis (TNM) stage, and lymph node metastasis, which suppressed breast cancer (BC) cell tumorigenesis. 17 Xiao et al 18 suggested that miR-22-3p promotes bladder cancer chemoresistance, which may serve as a new prognostic biomarker for bladder cancer patients. In addition, miR-22-3p is reported to play an essential role in the regulation of HCC progression. 19 In the ceRNA network, PPP1R9A was a target of hsa-miR-22-3p, and AC005523.2 may act as the sponge of hsa-miR-22-3p to capture PPP1R9A in HCC recurrence after LT.   DCUN1D1 is a transcription factor, which is increased significantly and associated with progression and prognosis of prostate cancer. 20 MBNL3 was identified to regulate cell invasion of pancreatic ductal adenocarcinoma. 21 In addition, MBNL3 was reported to promote tumorigenesis and indicate poor prognosis of HCC patients. 22 Huang et al 23 reported that miR-204-3p was involved in colorectal cancer metastasis. Decreased miR-204-3p was detected in gastric cancer specimens when compared with non-tumor specimens. 24 Cui et al 25 revealed that miR-204-3p inhibits the growth of HCC tumor endothelial cells. In the ceRNA network, MBNL3 and DCUN1D1 were targets of hsa-miR-204-3p, and AC005523.2 and hsa_circ_0054853 may act as the sponges of hsa-miR-204-3p to capture MBNL3 and DCUN1D1 in HCC recurrence after LT.
The sex-determining region Y-box (SOX) family is an important group of transcription factors involved in tumorigenesis and cancer, which critically control cell fate and differentiation in cancer. SOX8 is a member of the SoxE group in the SOX family, which has been reported to function as an oncogene and involved in the progression of triple-negative breast cancer. 26 High expression of SOX8 was usually associated with a poor prognosis in colorectal cancer. 27 SOX8 was also overexpressed in chemoresistant patients with tongue squamous cell carcinoma and was associated with poor prognosis. 28 In addition, highly expressed SOX8 was detected in HCC, which promoted cellular proliferation and enhanced tumor growth in HCC. 29 Elevated hsa_circ_0007291 was detected in thymoma, which was associated with pathological immune disorder in thymoma. 30 In this study, hsa_circ_0007291, an upregulated DEcircRNAs that covered the most DEmRNAs, was co-expressed with SOX8, which may indicate that hsa_circ_0007291 was involved in HCC recurrence after LT by regulating SOX8.
However, there are still some limitations in this study. First, the sample size of sequencing is small, and a large number of samples need to be collected for subsequent verification. Second, the specific mechanism of the identified mRNA-, lncRNA-, and circRNA-associated recurrence in HCC patients after LT is still unclear, and a large number of experiments are needed to verify it.

CONCLUSIONS
In conclusion, this study investigated the lncRNA, cir-cRNA, and mRNA expression profiles of tumor recurrence and without tumor recurrence in HCC patients and found altered lncRNA, circRNA, and mRNA expression pattern during the recurrence HCC after LT. Our results reveal key genes circRNAs and lncRNAs associated with recurrence in HCC patients after LT and lay the foundation for understanding the molecular basis of recurrence in HCC patients after LT. The precise mechanisms of the potential biomarkers in recurrence HCC after LT need to be confirmed by further validation or experiments.
Availability of Data: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Due to the need to continue the experiment in the later stage, the data will not be released for the time being.
Ethics Committee Approval: This study was approved by the ethics committee of Third Hospital of Hebei Medical University (20210101).
Informed Consent: All samples were collected after obtaining written informed consent from every participant.