Elevated expression patterns of P-element Induced Wimpy Testis (PIWI) transcripts are potential candidate markers for Hepatocellular Carcinoma

1.Introduction

Liver cancer continues to be a global health problem, with global rates increasing [1, 2]. By 2025, the disease is predicted to affect over 1 million people [3]. Hepatocellular Carcinoma (HCC) is the most common type of primary liver cancer, accounting for around 90% of cases. HCC is seen as a problematic health problem in Egypt for example, with the number of patients doubling over the last decade [4]. Several risk factors are recognised for the development and progression of HCC, but the most prominent ones are viral infections, i.e., Hepatitis B virus (HBV) and hepatitis C virus (HCV). Cirrhosis is also considered at a risk factor developing HCC [3]. HCC aetiology is additionally correlated with mutational changes ascribed from exposure to tobacco and aristolochic acid (AA), and non-alcoholic steatohepatitis (NASH), all of which were determined as probable pathogenetic cofactors in HCC [5]. There is a widely accepted protocol for diagnosing chronic liver disorders (CLD), which involves evaluating liver function using a series of serum-level enzyme assays and a significant tumour marker, α-fetoprotein (AFP); and imaging techniques such as ultrasound, computed tomography (CT), and magnetic resonance imaging (MRI), which have advanced dramatically in recent years. However, AFP has exhibited suboptimal results in terms of therapeutic surveillance and early identification [6]. Subsequently There is a need for more accurate serum biomarkers with increased sensitivity and specificity that complement AFP and improve clinical outcomes for patients.

Recent studies have shown that P-element-induced wimpy testis (PIWI) proteins could be used as markers for their diagnostic and prognostic values [7, 8]. Early Carcinogenesis has been linked to multiple epigenetic abnormal events such as; global hypomethylation of DNA, post-transcriptional changes of histones, dysregulation of noncoding RNAs (ncRNAs), and reactivation of transposable elements (TE) [9, 10, 11, 12, 13, 14]. PiRNAs (P-element induced wimpy testis (PIWI)-interacting RNAs) are short single-stranded ncRNAs, typically 25–33 nucleotides, and interact with PIWI proteins of the Argonaute family. PIWI proteins are involved in the synthesis of piRNAs and assemble ribonucleoproteins known as PiRNA-induced silencing complexes (pi-RISCs) in the cytoplasmic perinuclear foci or ‘nuage’, the mechanism operates at the transcriptional and post-transcriptional stages, and is based on complementarity with short RNA strands (piRNAs, miRNAs, and siRNAs). PIWI proteins and piRNAs were first thought to be implicated in germline and stem cells, with involvements in development, gametogenesis, proliferation differentiation, and maintenance of its integrity and stability via the inhibition of transposable elements’ (TEs) activation [15, 16, 17, 18, 19]. Their roles were also identified for self-renewal, fertilisation, organogenesis and epigenetic activation, expression of genes and proteins, maturation and plasticity of the brain, pancreas functions, and even fat metabolism [20, 21]. Emerging evidence has found their role in carcinogenesis and are related with prominent cancer hallmarks [5]. In humans, the PIWI protein family consists of four proteins: PIWIL1/HIWI, PIWIL2/HILI, PIWIL3, and PIWIL4/HIWI2 [22], belong to the class of cancer/testis antigens (CTAs), and their dysregulation is associated with cancer cell maintenance of proliferative signalling, apoptosis, stemness, genomic integrity, activating invasion, metastasis, mediating genomic instability, and boosting cell growth, to mention a few [23]. The abnormal expression of PIWIs in cancer were first discovered in 2011 [24] and the molecular mechanisms underlying PIWI’s oncogenic actions are controversial. Studies have also supported that PIWI proteins can be utilized cancer prognosis, and in combination with piRNAs they could also be employed for diagnosis [25]. Overexpression of PIWIL1/HIWI gene is seen in various cancers, including seminoma cell hyperplasia [25], oesophageal squamous cell carcinoma, gastric cancer [26] and pancreatic adenocarcinoma [27]. PIWIL2 gene variants transcribed by intragenic promoters, and shorter mRNAs were implicated in various cancers due to their carcinogenic characteristics. Thus, altered PIWI proteins and their variations seen in somatic malignant tumours may serve as diagnostic and prognostic biomarkers and therapeutic targets [28, 29]. Due to inconsistent findings in the literature, we examined the expression levels of the four human members of the PIWI family, both RNA levels by quantitative RT-PCR, in HCC patients (n= 50) and associated them with their clinicopathological characteristics.

2.Methods

2.7Statistical

All statistical analyses were performed using statistical software SPSS (Statistical Package for Social Science) statistical program version 21.0. A Power test, indicated that the standard deviation of control is 0.8 and the standard deviation for the regression errors will be 1.9. Regression was at 1.1, and that 50 study subjects and 25 normal controls will be an appropriate representation via regression with a probability of 85%, and Type I error was 0.05. as adapted from HCC molecular marker research [33]. Normality tests determined continuous variables, described as mean ± standard deviation (SD), or median and interquartile range (IQR) according to their distribution. Frequencies and percentages were used for categorical variables. A p value of < 0.05 was considered statistically significant. Mann-Whitney U was used, to compare means, Fisher’s exact test was used to determine the distribution of categorical variables between groups. Receiver operating characteristic (ROC) curve was used to generate the Cut-off values, area under the curve (AUC), sensitivity, and specificity for diagnostic potential. Odd ratios (OR), and 95% confidence intervals (CIs) were calculated by logistic regression to assess the relative risk for PIWIL mRNA.

Table 1

Demographics and Clinico-pathological characteristics for HCC patients

Clinico-pathological characteristics	Total number of patients N= 50 (%)
Age (Mean ± SD)	57.2 ± 8.1
Sex
Female	22 (44.0)
Male	28 (56.0)
ALT
7–55 U/L	61.4 ± 15.5
AST
8–33 U/L	65.8 ± 16.6
Alb
3.4–5.4 g/dL	2.3 ± 1.0
Bilirubin
0.3–1.2 mg/dL	2.9 ± 1.1
AFP
0 ng/mL–40 ng/mL	75.0 (40.0–150.0)
< 400 ng/mL
S. Creatinine
M: 0.74–1.35 mg/dL	0.9 ± 0.4
F: 0.59–1.04 mg/dL
No. of masses	1.1 ± 0.2
Tumour size	2.25 (0.75–4.25)
Tumour grade
I	29 (38.7)
II	11 (14.7)
III	10 (13.3)
Pattern
Acinar	30 (60.0)
Solid	17 (34.0)
Acinar/Solid	3 (6.0)
Steatosis	0.02 (0.02–0.04)
Stage
Fibrosis	8 (16.0)
Cirrhosis	42 (84.0)
HAI
A1	15 (30.0)
A2	35 (70.0)
A3	0 (0.0)
Hepatomegaly
Negative	41 (82.0)
Positive	9 (18.0)
Ascites
Negative	30 (60.0)
Positive	20 (40.0)
Splenomegaly
Negative	21 (42.0)
Positive	29 (58.0)
Oedema lower limbs
Negative	37 (74.0)
Positive	13 (26.0)

Alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin (Alb) and alpha-fetoprotein (AFP). No. of masses are represented as Mean and SD. But Alpha feto-protein, Tumour size, and Steatosis (Fatty degeneration of hepatocytes (% of cells)) are represented as Median and Interquartile Range IQR (25%–75%). While Sex, Grade, Pattern, Stage, HAI (Hepatitis Activity Index (grade of hepatitis), Hepatomegaly, Ascites, Splenomegaly, and oedema Lower Limbs are represented as Frequency and percent.

Figure 1.

Relative expression representation for PIWIL mRNA in HCC (N= 50), where: A) PIWIL1 mRNA in Serum, B) PIWIL1 mRNA in Tissue, C) PIWIL2 mRNA in Serum, D) PIWIL2 mRNA in Tissue, E) PIWIL3 mRNA in Serum, F) PIWIL3 mRNA in Tissue, G) PIWIL4 mRNA in Serum, & H) PIWIL4 mRNA in Tissue.

3.Results

3.1Patient characteristics

A total of 50 HCC participants were recruited in this study, the patients included 28 males (56%) and 22 females (44%), with a mean age of 57.2 ± 8.1 and 25 healthy individuals with no history of liver disease or alcohol consumption were included as controls. Individual demographic and clinical data of the studied groups are shown in (Table 1). The results for biochemical parameters were represented as mean ± SD for these tests, and had values of 61.4 ± 15.5, 65.8 ± 16.6, 2.3 ± 1.0, 2.9 ± 1.1, 0.9 ± 0.4 respectively, AFP however was 75.0 (40.0–150.0). By computed tomography, the number of tumour masses was detected with mean of 1.1 ± 0.2, while tumour size and steatosis were found at 2.25 (0.75–4.25) and 0.02 (0.02–0.04) respectively. Regarding pathological diagnosis, 29 (38.7%) were diagnosed GI, 11 (14.7%) were with GII and 10 (13.3%) were GIII, 30 (60%) were with acinar pattern tumour, 17 (34%) with solid tumour and only 3 patients (6%) were with acinar/solid tumour. For tumour staging 8 patients (16%) were fibrotic with the predominant number for the cirrhotic patients 42 (84%). Histopathological examination was performed using METAVIR scoring system to determine the hepatitis activity index (HAI),

Table 2

Clinico-pathological characteristics for HCC patients and their association with PIWIL mRNA expression

Clinico-pathological parameters	No. of cases N= 50 (%)	Tissue expression				Serum expression
		PIWIL1	PIWIL2	PIWIL3	PIWIL4	PIWIL1	PIWIL2	PIWIL3	PIWIL4
Tumour Grade		0.2	0.7	0.4	0.2	0.8	0.9	0.9	0.2
I	29 (38.7)	6.2 (4.4–13.4)	6.8 (3.5–25.8)	6.0 (3.5–9.0)	1.7 (0.8–2.9)	71.0 (18.2–100.3)	29.6 (24.8–41.6)	55.3 (29.0–122.5)	32.7 (17.0–97.8)
		P< 0.05*	0.9	0.5	0.5	0.1	0.6	0.9	0.5
II	11 (14.7)	28.0 (5.9–36.5)	6.4 (4.6–9.9)	9.0 (3.0–20.0)	1.0 (0.6–7.8)	60.1 (18.2–120.2)	29.6 (16.8–37.5)	55.3 (36.0–170.0)	22.8 (11.8–39.7)
		P< 0.08	0.9	0.9	P< 0.04*	0.2	0.7	0.3	0.2
III	10 (13.3)	10.1 (5.8–30.0)	6.2 (4.6–8.3)	7.0 (2.8–20.0)	0.7 (0.4–2.6)	100.3 (49.3–128.2)	30.1 (22.9–39.5)	189.7 (32.5–236.3)	39.7 (15.7–182.9)
Pattern		0.3	0.6	1	0.5	0.6	0.3	0.6	0.7
Acinar	30 (60.0)	7.9 (5.1–13.5)	6.4 (3.8–25.8)	6.0 (3.0–11.8)	1.7 (0.6–2.2)	65.6 (18.2–90.4)	30.1 (24.9–39.5)	55.3 (37.5–179.8)	34.9 (17.0–84.9)
		0.3	0.3	0.5	0.4	0.7	0.5	0.9	0.8
Solid	17 (34.0)	7.9 (4.1–28.8)	6.1 (4.1–6.7)	9.0 (3.0–20.0)	1.0 (0.5–5.9)	71.5 (18.2–152.2)	29.6 (8.9–35.3)	55.3 (22.0–189.7)	22.8 (11.8–55.9)
		0.4	P< 0.03*	0.2	0.2	0.6	0.1	0.3	0.1
Acinar/Solid	3 (6.0)	31.3 (6.2–31.3)	10.9 (9.9–10.9)	20.0 (7.0–20.0)	6.8 (1.2–6.8)	80.4 (3.1–80.4)	37.5 (0.9–37.5)	75.0 (22.0–75.0)	32.7 (11.8–32.7)
Stage
Fibrosis	8 (16.0)	0.7	0.5	0.4	0.8	P< 0.02*	0.8	0.7	0.9
Cirrhosis	42 (84.0)	7.9 (5.3–12.9)	8.3 (3.5–25.8)	6.0 (3.3–9.0)	1.9 (0.5–3.5)	8.0 (4.9–28.0)	6.2 (4.2–14.1)	8.0 (3.0–20.0)	1.4 (0.6–3.1)

The studied genes are represented as Median and Interquartile Range IQR (25%–75%); the data were analysed by Mann-Whitney U test. ^{*p <} 0.05, ^**p< 0.001. Grade, Pattern, Stage, Hepatomegaly, Ascites, Splenomegaly, and oedema Lower Limbs are represented as Frequency and percent.

Table 3

Clinico-pathological characteristics for HCC patients and their correlation with PIWIL mRNA expression

Clinico-pathological carameters	No. of cases N= 50 (%)	Serum								Tissue
		PIWIL1		PIWIL2		PIWIL3		PIWIL4		PIWIL1		PIWIL2		PIWIL3		PIWIL4
		r	p. value	r	p. value	r	p. value	r	p. value	r	p. value	r	p. value	r	p. value	r	p. value
Age (Mean ± SD)	57.2 ± 8.1	0.178	0.514	-0.289	0.042*	0.35	0.41	0.132	0.204	0.934	0.01*	0.89	0.031*	0.93	0.615	0.385	0.006*
Sex
Female	22 (44.0)	0.031	0.15	0.047	0.741	0.207	0.173	0.027	0.852	0.172	0.233	0.142	0.441	0.155	0.427	0.085	0.061
Male	28 (56.0)	0.053	0.24	0.072	0.35	-0.253	0.106	0.099	0.506	-0.173	0.245	-0.074	0.631	-0.129	0.387	-0.093	0.545
ALT
7–55 U/L	61.4 ± 15.5	0.03	0.838	-0.158	0.274	-0.078	0.611	0.087	0.546	-0.04	0.784	-0.02	0.891	-0.051	0.725	-0.085	0.558
AST
8–33 U/L	65.8 ± 16.6	0.015	0.917	-0.118	0.414	-0.118	0.439	0.022	0.88	-0.135	0.348	-0.058	0.688	-0.179	0.213	-0.193	0.179
Alb
3.4–5.4 g/dL	2.3 ± 1.0	0.048	0.742	0.076	0.601	0.049	0.748	-0.061	0.672	0.091	0.528	0.097	0.503	0.088	0.542	0.151	0.296
Bilirubin
0.3–1.2 mg/dL	2.9 ± 1.1	0.031	0.829	0.047	0.745	0.207	0.173	0.027	0.852	0.172	0.233	0.142	0.324	0.115	0.428	0.165	0.252
AFP
0 ng/mL–40 ng/mL	75.0 (40.0–150.0)	-0.057	0.703	-0.027	0.86	-0.253	0.106	0.099	0.506	-0.173	0.245	-0.074	0.623	-0.129	0.387	-0.093	0.534
< 400 ng/mL
S. Creatinine
M: 0.74–1.35 mg/dL	0.9 ± 0.4	-0.078	0.591	-0.072	0.62	0.037	0.81	0.183	0.204	0.034	0.814	0.03	0.835	-0.073	0.616	-0.134	0.352
F: 0.59–1.04 mg/dL
No. of masses	1.1 ± 0.2	-0.043	0.767	-0.102	0.481	-0.124	0.415	0.136	0.348	-0.131	0.366	-0.022	0.879	-0.128	0.377	-0.081	0.578
Tumour size	2.25 (0.75–4.25)	-0.043	0.782	-0.091	0.556	-0.067	0.685	0.064	0.682	-0.143	0.354	-0.126	0.416	-0.166	0.281	-0.205	0.182
Steatosis	0.02 (0.02–0.04)	-0.158	0.273	-0.028	0.848	-0.139	0.336	-0.189	0.189	-0.058	0.69	-0.049	0.738	-0.085	0.559	-0.086	0.554

Alanine aminotransferase (ALT), aspartate aminotransferase (AST), albumin (Alb) and alpha-fetoprotein (AFP). No. of masses are represented as Mean and SD. But Alpha feto-protein, Tumour size, and Steatosis (Fatty degeneration of hepatocytes (% of cells)) are represented as Median and Interquartile Range IQR (25%–75%). Sex, represented as Frequency and percent r= Pearson Correlation (2-tailed); ^{*p <} 0.05, ^**p< 0.001.

15 (30%) patients were rated as AI, 35 (70%) were A2 while none of the patients were A3. Abdominal ultrasounds detected hepatomegaly in 9 (18%) patients, ascites in 20 (40%) patients and splenomegaly in 29 (58%) patients and finally 13 (26%) out of the 50 HCC patients were detected with oedema lower limbs (Table 1).

Figure 2.

ROC curves of the studied PIWIL mRNA in (a) serum and (b) tissue samples. (C & D) are combined curves for serum and tissue, respectively. (E) ROC curve for AFP measured by ELISA in serum samples of HCC patients. Graphs depict the diagnostic performance of all PIWIL mRNAs in terms of specificity & sensitivity.

Table 4

ROC curve analysis for diagnostic performance for each of the studied PIWIL mRNAs in serum and tissue samples

	Groups	Cut-off	Sensitivity	Specificity	AUC	95% C.I		p value
						Lower bound	Upper bound
Serum	PIWIL1	< 1.6	100.0	100.0	1.0	1.0	1.0	< 0.001**
	PIWIL2	< 0.56	100.0	100.0	1.0	1.0	1.0	< 0.001**
	PIWIL3	< 8.87	100.0	100.0	1.0	1.0	1.0	< 0.001**
	PIWIL4	< 1.02	100.0	100.0	1.0	1.0	1.0	< 0.001**
	ELISA/AFP	> 0.97	84.0	64.0	0.828	0.746	0.910	< 0.001**
	Serum	Combined	100.0	100.0	1.0	1.0	1.0	< 0.001**
Tissue	PIWIL1	< 4.76	80.0	72.0	0.80	0.71	0.89	< 0.001**
	PIWIL2	< 3.69	84.0	64.0	0.80	0.71	0.89	< 0.001**
	PIWIL3	< 4.5	68.0	84.0	0.79	0.70	0.88	< 0.001**
	PIWIL4	< 0.94	64.0	68.0	0.66	0.56	0.77	< 0.01*
	Tissue	Combined	70.7	68.0	0.733	0.683	0.783	< 0.001**

Sn: Sensitivity, Sp: Specificity, AUC Area under curve and C.I: 95% Confidence Interval. ^{* p<} 0.05, ^** p< 0.001.

4.Discussion

PIWIL1, PIWIL2, PIWIL3 and PIWIL4, are designated as catalytic elements of the pi-RISCs complexes, the have implications in piRNAs’ biogenesis on the basis of complementarity [35, 36, 37]. Their roles vary at transcriptional and post-transcriptional epigenetic regulation. PIWIL1-2-3-4 RNAs were strongly expressed in HCC tissue, and circulating sera compared to the controls. The unexpected role of the PIWI-piRNA pathway has led to distinct functions of human PIWI proteins and mRNA in various cancer types [38, 39, 40]. Their involvement in multiple cancer hallmarks, has led to a possible representation as diagnostic and prognostic biomarkers [41, 42]. There are conflicting statements on the expression patterns of PIWIL1/PIWIL2/PIWIL3/PIWIL4, their prognostic, and predictive values, in addition to a complete absence for hepatocellular carcinoma in terms of mRNA transcript accumulation, and associated piRNAs, with recent studies on identifying novel piRNAs [43, 44, 45].

In terms of demographic data for the patients of the current study, the mean age was 57.2 ± 8.1, 44% of which are females and 56% males, typical for the risk factors associated with HCC (Table 1) and all typical parameters are represented within the table. We found a strong correlation between age and tissue expression for PIWIL1, 2, and 4, but not PIWIL3, and other clinical data HAI, hepatomegaly, ascites, splenomegaly, oedema lower limbs, number of masses was not significant. The aforementioned findings were in alignment with Taubert et al. and Al-Janab et al. who observed similar outcomes for no association with clinicopathological features apart from age for colorectal cancer and renal cell carcinoma respectively [46, 47], but this was refuted by Zhang et al. and Li et al. in breast and colorectal cancer respectively [41, 48]. For tumour grade, PIWIL1

Figure 3.

(A) miRNet prediction of PIWIL and HCC. (B) Reported cases with mutated PWIL genes in Liver cancer (C) Frequency of mutated PIWIL genes and their predicted variants’ implication for HCC through TCGA; via BioPortal for Cancer Genomics (https://www.cbioportal.org).

Figure 4.

(A) Genemania generated Network for PIWIL1, PIWIL2, PIWIL3, PIWIL4 and function predictions (B) Metascape results for Enriched ontology clusters. (C) PIWIL Subset network with force-directed clustering (D) Enriched network according to statistical significance, the darker the color, the more statistically significant the node is. (E) protein-protein interactions’ network from PIWIL genes. (F) Results for MCODE algorithm to identify neighborhoods where proteins are densely connected.

& PIWIL4 had significant difference with tissue expression, which was confirmed in a study by Litwin et al. for PIWIL1, & PIWIL2 in breast cancer [49].

Recently, several reports have indicated that aberrant expression of PIWI at the mRNA and protein levels occurs in various types of tumours [50, 51]. PIWIL1/HIWI was previously linked with several types of cancers, with a pattern of overexpression, moreover, it was correlated with tumour grading and staging [51, 52]. Our findings confirmed the apparent role of PIWIL mRNAs for HCC, with a pattern observed of overexpression for all PIWIL isoforms. Previously, both PIWIL1 & PIWIL2 was identified as overly expressed in colorectal, prostate, breast, cervical, gastric and bladder cancer [51], PIWIL1, 2 & 4 was observed as downregulated in renal cell carcinoma [50]. Meseure et al. conducted an investigation on the bio-pathological significance of the PIWI-PiRNA pathway through PIWIL1-2-3-4 mRNA expression levels in a panel of normal tissues and corresponding malignant tumours, and detected variable levels of expression across malignancies [53]. In addition, the relative expression of PIWIL2 mRNA was previously found to be higher in HCC tissues compared with adjacent normal liver tissues. A positive correlation was found between PIWIL2 expression and piR-Hep1 level according to Pearson’s correlation analysis [37, 38]. PIWIL2 acts as an oncogene by activating the STAT3/Bcl-xl cell signalling pathway through endogenous RNAi mechanism, hence inhibiting cell apoptosis and promoting cell proliferation. To the best of our knowledge, our study is the first to report the expression levels of all PIWIL mRNAs in serum and tissue, and to suggest their possible roles as diagnostic and prognostic biomarkers, in addition to their correlation to HCC development.

The involvement of PIWI proteins was linked with multiple hallmarks of cancer including invasion, apoptosis evasion, metastasis and cell proliferation, as such they possess prospective diagnostic factors and biomarkers for cancer prognosis [25, 54]. Significant increased level of PIWIL1 was reported for colon, bladder, and hepatocellular carcinoma. Expression of the four members of the PIWI proteins was viewed as distinct in tumour tissue when compared with the adjacent non-tumorous tissue [22, 55]. PIWIL3 and PIWIL1 were assessed for relative expression levels by [56] in colorectal cancer, and had non-significant expression statistically. Among all PIWIL genes, those assessed for expression and correlated with overall survival and recurrence-free survival were PIWIL3 & PIWIL4, in invasive urothelial bladder cancer [49]. Erber et al. reported a limitation of their study was not assessing the expression of PIWIL1 and PIWIL2 at the mRNA level, and depending on Immunohistochemistry, but they reported higher expression levels [57]. In the current study, expression of all PIWIL was assessed using real-time PCR in HCC patients. The levels of mRNA transcripts expression are reported for the first time for PIWIL1, PIWIL2, PIWIL3, PIWIL4 in tumour and nontumorous adjacent tissue, and matching serum samples HCC patients and healthy controls, the expression was correlated with clinical data. When compared to adjacent non-cancerous tissues, we found a significantly elevated expression for PIWIL1, PIWIL2, PIWIL3, &, PIWIL4 in HCC samples (p< 0.001).

These findings are consistent with prior findings in colorectal cancer, in which PIWIL1 mRNA levels in non-cancerous tissue were low or undetectable, but were dramatically raised in malignant tissue [58]. Additionally, PIWIL1/HIWI was reported to have marked expression levels in HCC tissue, for patients who had undergone curative resection [42]. In breast cancer, PIWIL1 and PIWIL3 gene expressions were reported to be upregulated, whereas PIWIL2 and PIWIL4 were downregulated compared with normal breast tissue [53]. It is noteworthy to state that our research detected significant association between PIWIL1 & PIWIL4 expression with increasing tumour grade (p< 0.05). PIWIL1 mRNA expression in serum was associated with tumour stage (p< 0.05). PIWIL2 expression in tissue was associated with tumour pattern (p< 0.05). These findings are similar to those of previous studies, which found aberrant mRNA expression in the varying stages of multiple cancer types, inferring the role of PIWIL-mRNAs in cancer development [54, 55, 51, 59].

In terms of diagnostic performance, our findings showed that serum had an overall sensitivity and specificity of 100%, and an AUC of 1.0, in comparison to normal serum, which shows an indication of disease prevalence but should be adapted for clinical settings i.e., patients with chronic liver diseases. Tissue samples exhibited sensitivity of 70%, specificity of 68%, and AUC of 0.733. Both of which were significant p< 0.001. Referring to impact on survival, PIWIL1, PIWIL4, PIWIL2 were reported for survival data, but not PIWIL3 (Krishnan et al., 2016). Research on the prognostic implications of PIWIL and PiRNAs was determined with higher sensitivity e.g., 83.3% sensitivity and 89.3% specificity for colorectal cancer (CRC) tissue expression in CRC patients [60, 61, 62, 63]. To assess the HCC risk level logistic regression was done and had a statistical significance (p< 0.001) in serum, and (p< 0.05) in tissue, these results are in agreement with Mai et al. for CRC cases diagnosed over the course of 3 years at 7.23, 2.80, 2.45, and 1.24, respectively [61] and similar findings were also reported [59]. Tosun et al. evaluated the predictive value of serum expression level of PIWIL2 mRNA and proteins in prostate cancer, with a strong correlation [64]. The reactivation of PIWI expression in cancer clearly suggests that these proteins are involved in the growth and differentiation of tumours [65]. According to Table 2, there was no clear correlation between the values of AFP and the expression of the PIWIL at the mRNA level for tissue and serum, the literature indicates that, serum AFP has a specificity of 76%–94%, and a sensitivity of 39%–65% for HCC [66]. Our ROC curve for ELISA/AFP was significant (p< 0.001), with sensitivity of 84%, Specificity of 64% (Table 4 and Fig. 2E), indicating better performance for PIWIL mRNA, in spite of a lack of disease specificity, and suggesting a panel use. Disruptions in PIWI-piRNAs pathway regulation has an effect on biological processes involving cancer progression, they comprise apoptosis, migration, proliferation and metastasis, which represents an indication that PIWI proteins and piRNAs can be used as diagnostic biomarkers, or novel therapeutic targets for the treatment of hepatocellular carcinoma [65], and that manipulation of gene expression for PIWI and PiRNAs could lead to more pronounced management in cancer progression, and an enhanced patient recovery [67].

PIWIL genes have several transcripts, some of which appear to be transcribed by putative intragenic promoters rather than a canonical promoter, which was associated with tumorigenesis [65]. PIWIL expression was revealed to have a direct predicted influence on HCC progression, through PIWIL2, and PIWIL4, this was found through two novel associations DDX4 (Fig. 4E), and miR-124-3p.1 (Fig. 3A), Recent studies have shown that downregulation of miR-124-3p.1 was associated with poor survival, early recurrence and sorafenib sensitivity in HCC patients [42], and our findings have demonstrated a novel direct interplay between PIWIL2, PIWIL4 and miR-124-3p.1, the pathway additionally indicates a several target miRNAs which could be used as possible therapeutic targets for PIWIL2 miR-33b-3p and miR-519d-3p were identified, while for PIWIL4, miR-129-2-3p, miR-200b-3p, and miR-212-3p were found, all of which are related to miR-124-3p.1, a microRNA which when upregulated, negative affects proliferation and migration in hepatocellular carcinoma via a phosphoinositide 3-kinase catalytic subunit alpha (PIK3CA) pathway [68]. and miR-101-3p was additionally detected for PIWIL4, and was reported to negatively influence HCC proliferation and metastasis through the HGF/c-Met pathway [70]. Thus, aberrant expression of PIWIL2 and PIWIL4 is a possible mechanism of tumour suppressor inactivation Fig. 3A, and the associated miRNAs could represent attractive therapeutic targets for combined therapies where a specific antibody for PIWIL2 or the PIWI/miRNA RISC complex could be targeted [71]. PIWIL expression is also shown have a direct predicted influence on HCC progression, through PIWIL2, and PIWIL4, through association with DDX4 (Fig. 4). Studies have shown that elevated levels of DDX4, indicates its ability to promote the stemness of breast cancer stem cells by regulating the expression of proteins such as Oct3/4 and Sox-2 and promoting disease progression [72]. In this regard, its upregulation or overexpression promotes proliferation, suggesting an oncogenic role, its association through the PPI network (Fig. 4), shows a novel interaction with PIWIL2 and PIWIL4 and further analysis into PIWIL2 and PIWIL4 expression and silencing is recommended.

5.Conclusion

Finally, PIWI mRNA, PIWI proteins, and piRNAs were identified as germline markers. DNA methylation, histone methylation, histone acetylation, and histone ubiquitination not only play significant transcriptional regulatory roles, but PIWI family proteins can break mRNA under the supervision of piRNA, suggesting a post-transcriptional regulatory function. However, the method of control of PIWI/piRNAs in cancer appears to be unique. Most studies have established that PIWI mRNA, proteins, and piRNA appear to govern tumours as two distinct entities rather than as a unified entity. As a result, it is critical to investigate how the PIWI protein regulates tumours independently of piRNA. The specific molecular biological mechanism behind the effect of PIWIL on the occurrence, progression, and prognosis of HCC is currently unknown and deserves additional investigation. Our findings provide a unique viewpoint on the activities of PIWIL at the mRNA level in HCC development, as well as a different pattern of overexpression that provides potential candidates for HCC disease progression and risk assessment. PIWIL mRNAs are overexpressed in HCC tissue and serum samples, the expression patterns could be valuable molecular markers for HCC, due to their association with age, tumour grade and pattern. To the best of our knowledge, our study is the first to report the expression levels of all PIWIL mRNA and to suggest their remarkable values as diagnostic and prognostic biomarkers, in addition to their correlation to HCC development. Additionally, a therapeutic opportunity might be also suggested through in silico miRNA prediction for HCC and PIWIL genes through DDX4 and miR-124-3p. The epigenetic regulation of the identified changes in PIWIL1, PIWIL2, PIWIL3, and PIWIL4 at the transcriptional and protein levels warrants additional investigation, which could have significant clinical relevance, and should be examined for therapeutic roles. A large sample size investigation will aid in analysing the link between PIWIL expression/co-expression and HCC prognosis. This can serve as the preliminary foundation for PIWIL as molecular markers of early-stage diagnostic and prognostic evaluation, as well as targeted cancer therapies.