Interaction between HOTTIP, H19, and HOTAIR long noncoding RNAs and miRNA‑152 in cases of HCC caused by HCV infection

Background Liver cancer (hepatocellular carcinoma “HCC”) remains a significant health issue without prompt detec‑ tion and appropriate prevention. By interacting with each other, long noncoding RNAs (lncRNAs) and microRNAs (miRNAs) can form gene regulatory networks. Specifically, we aim to determine whether the lncRNAs (HOTTIP, H19, and HOTAIR) and miRNA‑152 interact in a significant manner in the progression of Hepatitis C virus (HCV) patients to HCC. This is followed by the question of whether these biomarkers can be used to diagnose and prognose nonin‑ vasively. We used online computational techniques to predict which miRNA group is likely to affect the lncRNAs being examined. This study involved 133 participants. 103 patients with HCV were included in the study, which was divided into two groups: Group I, with 65 cases of chronic liver disease without HCC, and Group II, with 38 cases of chronic liver disease with HCC. In addition, 30 healthy volunteers served as controls. In this study, a qRT‑PCR was used to test gene expression. Results A consistent reverse correlation has been observed between lncRNAs and miRNA‑152 as the disease progresses. Conclusion According to our findings, the studied biomarkers may be useful as noninvasive biomarkers for prog‑ nosis in patients with HCV Genotype 4 who develop liver cirrhosis and HCC. Many miRNAs, including miRNA‑19a and miRNA‑106a, may interact with lncRNAs that have been investigated in addition to miRNA‑152.


Background
Among all cancers, hepatocellular carcinoma (HCC) causes the second-highest number of deaths globally [1].Several factors are related to chronic cirrhosis, including Obesity, alcoholism, chronic hepatitis B and C, and metabolic disturbances [2].The overall 5-year survival rate appears to be low at 18%, despite improvements in surgical procedures among HCC patients [3].In most cases, surgical resection or liver transplantation are not possible because of the late diagnosis.It has been reported that HCC has a recurrence rate of up to 50% within five years of surgery.The development of novel treatments for late-stage HCC may significantly improve the outcome of these patients [4].For this reason, it is imperative to diagnose HCC as soon as possible to obtain efficient and effective therapy and boost the probability of long-term survival.Understanding the genetic mechanisms involved in HCC development and progression is crucial to accurate detection at an early stage.It is also crucial to create effective approaches to treatments.
Several networks of genes and an imbalance of signaling pathways are hallmark features of HCC, as with many other cancer types [5,6].Genetic dysregulation affects proteins coding for genes, and non-coding RNAs (ncR-NAs) are also affected [7].There are approximately 98% of human genomes encoded by noncoding transcripts, the majority of which are processed into microRNAs and long noncoding RNAs (lncRNAs) [8].It is a general property of these RNAs that they can effectively carry out their biological roles [9].Newly discovered non-coding RNA (ncRNA) molecules have been proven to be essential players in a broad scope of biological and pathological processes [10,11].A number of ncRNAs have been shown to play a significant role in hepatocellular carcinoma pathogenesis [12,13].The conservation of lncR-NAs and miRNAs, for example, has been demonstrated over a long period of time.Due to this, it can be argued that they play a functional role indirectly, such as through their influence on a variety of biological and pathological processes, including tumor rates [14].
These RNAs affect cell proliferation, senescence, differentiation, stress responses, immune system activation, and apoptosis, in addition to the aforementioned two major classes of noncoding RNAs [15].Numerous medical conditions, including malignancies, neurological illnesses, and metabolic syndromes, are associated with noncoding RNAs.They serve a crucial function in gene expression, suggesting they may be viable therapeutic targets for diseases.In addition, they serve as biomarkers for disease diagnosis and prognosis.Consequently, research efforts have focused extensively on how noncoding RNAs interact with the regulatory network [16].
Along with the bioinformatics analysis we performed to select the lncRNAs (HOTTIP, H19, and HOTAIR), we found that this study summarizes seven well-documented lncRNAs in HCC: H19, HOTAIR, HULC, HOTTIP, MALAT1, MVIH, and MEG3.It was thought that upregulation of lncRNAs in HCC plays an oncogenic role [17].Moreover, a previous study addressed the role of HOT-TIP, H19, and HOTAIR in HCC progression in Egyptian HCV patients [18].Prior studies have shown that H19 [19], HOTAIR [20], and HOTTIP [21] are upregulated in the HCC tissue in comparison with the normal tissue.
A significant decrease in miR-152-3p expression was observed in HCC tissues as compared to non-tumor tissues [22].Furthermore, a previous study found that miRNA-152 was downregulated in HCC Egyptian patients with genotype 4 HCV infection [23].Therefore, analyzing the expression levels of these ncRNAs in the early stages of HCC can provide valuable insight into the disease progression and might be useful in early diagnosis and treatment.

Aim of the work
Our hypothesis is that these lncRNAs (HOTTIP, H19, and HOTAIR), as well as miRNA-152, are useful in detecting HCC in patients with HCV and predicting disease progression.We used online computational tools to assess interactions between miRNAs and lncRNAs.We also examined the association of these lncRNAs and miRNA-152 with clinical outcomes.Our final objective was to investigate the association between our candidate's expression quantities and disease progression.

Patients and methods
Following the guidelines of the Theodor Bilharz Research Institute's Research Ethics Committee (TBRI-REC) (FWA00010609), all patients were asked to provide informed consent according to the Helsinki Declaration of 1975.The TBRI-REC number is PT (673), which indicates that the local ethical committee approved this project.

Inclusion criteria
The Gastroenterology and Hepatology Section at the TBRI in Giza, Egypt, provided the data for the present study.The study took place over six months from April to October 2020.During this period, 103 patients were included, and their data regarding demographics, clinical diagnosis, and treatment were collected.
Patients were devided into two groups: chronic HCV liver disease (CLD) without HCC as group I (65 patients) and group II chronic HCV CLD with HCC (38 patients).Group I was further subdivided according to the presence of liver cirrhosis into two subgroups, group 1a, those without liver cirrhosis (30), and group 1b, those with liver cirrhosis (35) respectively.Moreover, we recruited 30 healthy individuals of matched gender and age as controls.
All patients and controls were subjected to a thorough history and comprehensive clinical examination.
All HCV patients included had persistent viremia for the last six months and did not receive HCV treatment with direct-acting antiviral agents (DAAS) or interferon.
Liver cirrhosis was diagnosed with characteristic abdominal ultrasound findings: coarse echotexture, an irregular border liver, and a shrunken liver.Those with focal lesions are confirmed by triphasic CT abdomen for HCC diagnosis.Additionally, HybProbe probes are used to genotype HCV.

Exclusion criteria
Based on the inclusion criteria, the study excluded patients with Schistosomiasis, dual Hepatitis B and C infections, nonalcoholic steatohepatitis, another chronic viral infection, biliary disorders, autoimmune hepatitis, non-HCC cancer patients, hepatotoxic drugs, alcoholism, diabetes, or patients with HCV treated with DAAS or interferon.

Sampling and storage
Under strict aseptic conditions, about 7 ml of peripheral venous blood were collected using vacuum blood collection tubes by clean venipuncture and distributed as 2.5 ml in EDTA tubes as a complete blood picture, 2.5 ml in another sterile EDTA tube (stored at −80 °C) as a viral RNA extract for HCV genotyping, miRNA, and mRNA extraction.Furthermore, 2 ml were centrifuged at 3000 rpm for 10 min after clotting at 37 °C, and the collected serum was stored at −80 °C for liver, kidney, and other specific serological tests.

Molecular analysis 2.4.1 Bioinformatics analysis
In the analysis of bioinformatics, we selected H19, HOT-TIP, and HOTAIR as targets for miRNA-152 based on an integration of online databases: miRNet version 2.0, a database derived from the following meticulously documented databases: miRecords and miRTarBase v8.0, as well as a gene prediction tool for lncRNA target genes (miRnet) and miRNA-centric network visual analytics platform.

HCV genotyping
The HCV genotyping was performed using the HybProbe probes for PCR Cat.No. (03003248001), which can be found at (https:// www.roche.com).

Expression of the targeted gene
Extraction of total RNA was performed using a highpurity RNA isolation kit (version 12, 2011), Cat.No: (11828665001, ROCH).For gene expression determination, RNA SYBR green I (Lightcycler EvoScript, 2017) master mix, Roche, Cat No: (07800134001), was used.Table 1 contains the primer sequences.The results of the lightCycler RNA SYBR green I master are to be analyzed using the combination of the SYBR green I filter (465-510).The CT approach was used to analyze this study.MiRNA quantity and quality between samples Table 1 Oligo nucleotides seq.used in RT-PCR are compensated by using the housekeeping gene U6 [24][25][26], while Β-actin (used as an endogenous control to normalize the amount of total mRNA in each sample) of HOTTIP, H19, and HOTAIR between different samples.Comparisons of gene expression levels between test and control samples were performed (served as a calibration sample) according to the formula 2 −ΔΔCT .

Statistical analysis
For the analysis of the data, Microsoft Excel 2016 and IBM SPSS Statistics for Windows, version 26, were used (IBM Corp., Armonk, New York, USA).For non-normal variables, a 0.05 p value was determined as statistically significant when the 25th and 75th percentiles were present.Statistical analysis was conducted using the Mann-Whitney U test for non-normal variables.In order to assess the diagnostic performance of the studied genes, receiver operating characteristic curves (ROC) were calculated.ROC curves and their areas under ROC curves (AUC) are accurate metrics for assessing the prediction accuracy of selected tests.Study groups were diagnosed based on the point with the greatest combination of sensitivity and specificity.We used bivariate Pearson correlation (r) to analyze the relationship between the markers under examination.The performance of prognosis and/ or risk assessment was analyzed using binary logistic regression analysis.

Interactions and bioinformatics analysis perspectives
In Fig. 1, using the miRNet web server database, it can be seen that HOTTIP, H19, and HOTAIR, along with miRNA-152 and all functionally targeted miRNAs, influence one another in multiple ways.In this study, miRNA-106a and miRNA-119a were presented as targeted miRNAs.Particularly, these miRNAs are closely associated with HOTAIR, H19, and HOTTIP, which are associated with HCCs with different expression variants.In addition, this could reveal new relationships between them as well as potential associations with HCC development.

Gene expression
It is evident that the expression of HOTTIP, H19, and HOTAIR is consistently increased, while the expression of miRNA-152 is consistently decreased, according to the fold change principle (Table 2) (Fig. 2).The expression values of HOTTIP (p 0.001), H19 (p 0.001), and HOTAIR (p 0.01) were statistically different from controls.Compared to noncirrhotic patients, HCC patients and patients with cirrhosis expressed significantly higher levels of HOTTIP (p 0.03), H19 (p 0.04), and HOTAIR (p 0.001).These values were also higher among patients with HCC when compared with patients without cirrhosis (p values of 0.001, 0.001, and 0.001).There was a significant increase in HOTTIP (p 0.001), H19 (p 0.013), and HOTAIR (p 0.005) expression in patients with HCC compared with patients with cirrhosis (Table 2, Fig. 2a, b,  c).
The expression of miRNA-152, however, decreased significantly (p 0.04) in all patient groups in comparison to the control group.Compared to non-cirrhotic patients, the expression of miRNA-152 was lower in cirrhosis patients (p 0.007) and HCC patients (p 0.001).MiRNA-152 expression was significantly lower in patients with HCC cases (p 0.001) compared to those with cirrhosis (Table 2, Fig. 2d).
It is also noteworthy that there is a direct positive correlation between the HOTAIR and HOTTIP genes (r = 0.813, p 0.001).Additionally, there is a significant direct positive correlation between the HOTAIR gene and the H19 gene (r = 0.349, p 0.001).Moreover, H19 has a positive correlation with HOTTIP (r = 0.315, p 0.001).

Diagnostic accuracy
Utilizing ROC curves, it was evaluated how successfully the tested groups could be diagnosed by the HOT-TIP, H19, HOTAIR, and miRNA-152 genes.Neither the HOTTIP nor the HOTAIR genes can be used to distinguish between cirrhotic and non-cirrhotic cases.In contrast, H19 showed an AUC of 0.77, p 0.0001, while miRNA-152 showed an AUC of 0.7, p 0.007 (Table 3, Fig. 4a).
These findings suggest that these four biomarkers can distinguish between HCC and cirrhotic patients with a reasonable degree of sensitivity, specificity, and AUC.In addition, their respective cut-off values can be utilized for accurate patient diagnosis.

Prognostic performance
Univariate logistic regression analysis determined that H19 and miRNA-152 were prognostic and/or predictive factors for cirrhosis progression.The possibility of developing cirrhosis increases by factors of 1.36 (p 0.02) and 2.32 (p 0.01) as H19 and miRNA-152 are both increased by one degree.On the basis of the HOTAIR and HOTTIP genes, patients with cirrhosis cannot be distinguished from those without cirrhosis (Table 4).

Discussion
In the absence of preemptive determination and appropriate treatment, hepatocellular carcinoma remains a significant health problem.Clinical diagnostics have successfully used a limited number of molecular biomarkers.HCC biomarkers have been extensively used as diagnostic or prognostic tools, as well as therapeutic targets.By interacting with microRNAs, messenger RNAs, and proteins, lncRNAs contribute significantly to cancer carcinogenesis and development [27].Furthermore, lncRNAs are implicated in tumorigenesis and cancer development based on numerous clinical observations Fig. 1 Outlook of all possible functional target miRNAs from the miRNet webserver database.According to hepatocellular carcinoma and HCV infection, the network appears to consist of all interacted miRNAs plus miRNA-152.Also, the interactions with targeted lncRNAs that could interact with them should they be exposed to a specific disease and experimental studies.In order to control gene expression, long noncoding RNAs and microRNAs work together.However, just a handful of lncRNA-miRNA interactions have been uncovered so far, and there is a lack of widespread access to computational approaches for predicting novel connections.Due to the intrinsic patterns in their interactions that lncRNAs and miRNAs share, it is possible to predict the underlying lncRNA-miRNA interactions from known ones The study examined the relationship between H19, HOTTIP, and HOTAIR as noninvasive biomarkers for identifying HCC accompanying genotype 4 HCV in patients with HCC on top of HCV genotype 4.
In our study, online computational methods were used to show that HOTTIP, H19, and HOTAIR, as well as miRNA-152 and all functionally targeted miRNAs, influence each other in multiple ways.The new miRNAs (miRNA-106a and miRNA-119a) also interacted with the lncRNAs we were interested in.A number of these miRNAs are closely associated with HOTAIR, H19, and HOTTIP, which are associated with HCCs with different expression variants.Additionally, this may reveal new relationships between them as well as potential associations with the development of HCC (Fig. 1).
There has been much research examining the relationship between miRNAs and lncRNAs, and more information is now becoming available regarding how miRNAs function in lncRNAs [8].Due to the role that miRNAs and long noncoding RNAs play in regulating gene expression and getting together to network, their interactions are strictly controlled.There is considerable evidence that lncRNAs and miRNAs are involved in many diseases, but further clarification is needed [28].Because of this, the new lncRNA-miRNA regulatory circuit HOTTIP, H19, Fig. 2 Relative gene expression of the studied genes in the studied groups: A HOTAIR gene expression in the studied groups.B HOTTIP gene expression in the studied groups.C H19 gene expression in the studied groups.D miRNA-152 gene expression in the studied groups.U6 was used as an endogenous control for miRNA, while Β-actin were used for HOTTIP, H19, and HOTAIR.Gene expression was calculated using three replicates for the fold change, while the fold change was calculated using the Law of Fold Change, which is 2 −ΔΔCT and HOTAIR-miRNA-152 could give patients with HCC on top of HCV a new epigenetic therapeutic target.
Gene expression results showed upregulation in H19, HOTTIP, and HOTAIR expression gradually with the progression of disease.At the same time, miRNA-152 levels are reduced in the studied patients when HOT-TIP, H19, and HOTAIR are upregulated, which is directly associated with disease progression (Table 2, Fig. 2).In addition, the correlation study showed an inversal correlation between the expression of miRNA-152 and H19, HOTTIP, and HOTAIR expression (Fig. 3).Based on the results of the mechanistic study, the expression of miRNA-152 is inversely related to the expression of HOTTIP, H19, and HOTAIR.The preceding relationship comes mostly in harmony with similar studies, which found that an lncRNA-miRNA-mRNA network is associated with recurrence in HCC and that lncRNA 'SNHG3' could promote the recurrence of HCC by regulating ASF1B expression via sponging of miR-214-3p [29].
A ROC curve was constructed to analyse the diagnostic accuracy of gene expression markers in cirrhotic individuals compared to non-cirrhotic individuals with varying cut-off points.Regarding H19 and miRNA-152, there was a significant difference in this study.The HOTTIP and HOTAIR lncRNA levels were not significantly different between cirrhotics and non-cirrhotics.That means HOTTIP and HOTAIR were unable to discriminate between cirrhotics and non-cirrhotics, which is contrary to H19 and miRNA-152.
HOTTIP, H19, HOTAIR, and miRNA-152 showed significant differences between patients with cirrhosis and HCC.All of them appear to be able to distinguish between patients with HCC and those with cirrhosis, according to the evidence.
Interestingly, HOTTIP, H19, HOTAIR, and miRNA-152 levels were significantly different between the HCC and CLD groups.In order to differentiate HCC patients from those with CLD generally, several factors have been demonstrated to be useful.A number of these factors can be identified, including HOTTIP, H19, HOTAIR, and miRNA-152.Our results are consistent with theirs, despite the fact that we used a different case study and strategy [30].Their findings indicate that lncRNAs, miR-NAs, and mRNAs are differentially expressed in normal and cancerous tissues.Additionally, altered lncRNAs and miRNAs may serve as potential diagnostic biomarkers for papillary thyroid carcinoma (PTC).
According to regression analysis, the expression levels of H19 and miRNA-152 were reliable prognosticators in patients with cirrhosis.The expression levels of HOTAIR and HOTTIP did not reach statistical significance, despite this fact.When comparing patients with cirrhosis or chronic liver disease to those with HCC, the levels of HOTTIP, H19, HOTAIR, and miRNA-152 were found to be strong predictors of an HCC diagnosis.According to [31], lncRNA-CASC7 was significantly upregulated in the serum of patients with HCC and closely associated with tumor number, IM, tumor size, and TNM stage, which could be used as a promising diagnostic biomarker.In a recent study [18], the serum levels of HOTTIP, H19, and HOTAIR were upregulated in Egyptian HCC patients with HCV genotype 4.Moreover, they can be used as noninvasive early-diagnosis biomarkers for HCC.Several potential biomarkers for HCC have recently been  identified through recent research.These results show that lncRNAs that are high may be useful molecular targets for treating HCC because they are turned down by miRNAs [17].Biomarkers for cardiac, neurological, and cancer disorders are commonly derived from both non-coding RNA groups [32][33][34][35][36][37], and tissue and liquid biopsies have been used to examine their potential as biomarkers for these conditions.It has been found that established protein biomarkers can be used in conjunction with miRNA and lncRNA-based tests to improve the specificity and sensitivity of the diagnosis [38].
In a later study, the visual analytics platform server on miRNet was used to look at the regulatory interactions between HOTTIP, H19, and HOTAIR as lncRNAs targeting miRNA-152.Other miRNAs may target all of these targets as well.This analysis identified all miRNAs that may target the lncRNAs studied in HCC development.There is a surprising finding from this study: the targeted miRNAs (miRNA-106a and miRNA-19a) are non-coding RNAs with a high profile of association with prostate cancer.Additionally, in the previous study [39], they discovered that miRNA-106a was associated with ovarian cancer-related EMT, while miRNA-19a was associated with metastases.
There has been growing evidence that miRNAs and lncRNAs function together to establish a regulatory network for genes.As a result, lncRNA and miRNA interactions provide valuable information regarding the molecular mechanisms underlying a number of complex diseases.It is the fact that only a few lncRNA-miRNA interactions are known that poses the biggest challenge (i.e., lncRNA-miRNA interaction networks are quite sparse), as well as the insufficient knowledge regarding the mechanisms underlying these interactions [8].The findings validate the theory we presented throughout this research and raise additional questions regarding how the remaining target miRNAs are concerned with tumor development.For a full Fig. 4 ROC curve of the studied genes in the studied groups.A ROC curve to discriminate between the cirrhotic versus non-cirrhotic groups.B ROC curve to discriminate between HCC versus the cirrhotic group.C ROC curve to discriminate between the HCC versus CLD groups.Receiver operating characteristic curves (ROC) were done to evaluate the diagnostic performance of the studied genes.ROC curves and their areas under ROC curves (AUC) are accurate metrics for assessing the prediction accuracy of selected tests.Study groups were diagnosed based on the point with the greatest combination of sensitivity and specificity understanding of their role in the development of HCC, more extensive research is required.

Conclusion
Accordingly, we can conclude from the above findings that lncRNAs (HOTTIP, H19, and HOTAIR) alter the expression of genes in patients with cirrhosis and HCC through their interaction with miRNA-152.Cancer disruption and development are caused by disruptions of the miRNA-152 pattern in HCC.Also, we may be able to provide reliable HCC-specific biomarkers based on the gene expression results for our candidates.Based on the results of these analyses, new therapeutic targets may be identified for treating HCV-induced cirrhosis and liver cancer.More interaction partners became apparent when computational tools were examined in more detail.There are other genes that may be present in the targeted domain of lncRNAs (HOTTIP, H19, and HOTAIR), in addition to miRNA-152.Consequently, patients with liver cirrhosis brought on by HCV and HCC would benefit from a larger, more in-depth pilot investigation that makes use of genome-wide transcriptional profiling and analytics based on bioinformatics.This study has some limitations, including a larger sample size and studying the association between different stages of cancer and these markers to assess whether these markers can be used for prognosis.Future studies should address these limitations by using a large sample size and a cross-sectional nature that would suggest correlation and need further confirmation by prospective studies.

Table 2
Gene expression of the studied genes in the studied groups The fold change results depend on the fold change low: Fold-Change (2 −∆∆CT ) is the normalized gene expression (∆CT) in the test sample divided the normalized gene expression (∆CT) in the control sample.(Fold-change values less than one indicate a negative or down-regulation) All parameters are represented as Median with Interquartile range (25-75%) of the fold change of the studied groups, the data were analyzed by Mann-Whitney U test *p value is significantly different comparing with CLD group (I) a p value is significantly different comparing with control group b p value is significantly different comparing with non-cirrhotic group c p value is significantly different comparing with Cirrhotic group 1 Initial p value < 0.05 is significant, 2 Initial p value < 0.01 is highly significant

Fig. 3
Fig. 3 Correlation study between miRNA-152 and the studied lncRNAs.A Correlation between miRNA-152 and HOTAIR.B Correlation between miRNA-152 and HOTTIP.C Correlation between miRNA-152 and H19.Bivariate Pearson correlation (r) to analyze the relationship between the markers under examination

Table 3
Diagnostic performances of the studied genesSn sensitivity, Sp specificity, AUC area under curve and CI: 95% confidence interval **p value < 0.01 is highly significant

Table 4
Prognostic performances of the studied genes OR odd ratio, CI confidence interval and p value of Prognostic viability were calculated depending on logistic regression analysis *p value < 0.05 is significant, **p value < 0.01 is highly significant