Correlations of Long Noncoding RNA HNF4A-AS1 Alternative Transcripts with Liver Diseases and Drug Metabolism

Abstract

Hepatocyte nuclear factor 4 alpha antisense 1 (HNF4A-AS1) is a long noncoding RNA (lncRNA) gene physically located next to the transcription factor HNF4A gene in the human genome. Its transcription products have been reported to inhibit the progression of hepatocellular carcinoma (HCC) and negatively regulate the expression of cytochrome P450s (CYPs), including CYP1A2, 2B6, 2C9, 2C19, 2E1, and 3A4. By altering CYP expression, lncRNA HNF4A-AS1 also contributes to the susceptibility of drug-induced liver injury. Thus, HNF4A-AS1 lncRNA is a promising target for controlling HCC and modulating drug metabolism. However, HNF4A-AS1 has four annotated alternative transcripts in the human genome browsers, and it is unclear which transcripts the small interfering RNAs or small hairpin RNAs used in the previous studies are silenced and which transcripts should be used as the target. In this study, four annotated and two newly identified transcripts were confirmed. These six transcripts showed different expression levels in different liver disease conditions, including metabolic dysfunction-associated steatotic liver disease, alcohol-associated liver disease, and obesity. The expression patterns of all HNF4A-AS1 transcripts were further investigated in liver cell growth from human embryonic stem cells to matured hepatocyte-like cells, HepaRG differentiation, and exposure to rifampicin treatment. Several HNF4A-AS1 transcripts highly displayed correlations with these situations. In addition, some of the HNF4A-AS1 transcripts also showed a strong correlation with CYP3A4 during HepaRG maturation and rifampicin exposure. Our findings provide valuable insights into the specific roles of HNF4A-AS1 transcripts, paving the way for more targeted therapeutic strategies for liver diseases and drug metabolism.

SIGNIFICANCE STATEMENT

This study explores the alternative transcripts of HNF4A-AS1, showing how their expression changes in different biological conditions, from various liver diseases to the growth and differentiation of hepatocytes and drug metabolism. The generated knowledge is essential for understanding the independent roles of different transcripts from the same lncRNA in different liver diseases and drug metabolism situations.

Introduction

Over the past three decades, long noncoding RNAs (lncRNAs) have been extensively investigated, revealing their important functions in many physiological processes, such as gene expression regulation (Sun et al., 2018; Statello et al., 2021), epigenetic regulation (Chen et al., 2020; Herman et al., 2022), serving as structural components (Bergmann and Spector, 2014; Engreitz et al., 2016), and involvement in alternative splicing (Tripathi et al., 2010; Gonzalez et al., 2015). LncRNAs are also implicated in multiple pathological conditions, including cancers (Huarte, 2015; Yang et al., 2022), autoimmune diseases (Zou and Xu, 2020), and metabolic disorders (Alipoor et al., 2021), highlighting their broad impact on human health. LncRNA hepatocyte nuclear factor 4 alpha antisense RNA 1 (HNF4A-AS1) has gained attention for its dysregulation in hepatocellular carcinoma (HCC) and other diseases like Crohn's disease and lung adenocarcinoma (Esposti et al., 2016; Haberman et al., 2018; Zhang et al., 2020). Its critical role in modulating cytochrome P450 (CYP)-mediated drug metabolism, especially affecting key enzymes such as CYP1A2, 2B6, 2C9, 2C19, 2E1, and 3A4, has made it stand out in pharmacological research (Chen et al., 2018; Wang et al., 2021). For instance, HNF4A-AS1 impacted ritonavir-induced hepatotoxicity by affecting the binding of PXR and histone modification status in the CYP3A4 promoter, changing the CYP3A4 expression level (Wang et al., 2022). Also, HNF4A-AS1 affected acetaminophen-induced liver injury mainly through altering CYPs in HepaRG cells (Chen et al., 2020). These findings suggest that HNF4A-AS1 RNA could be a promising target for studying the roles of lncRNAs in liver diseases and drug metabolism.

Several transcripts are annotated from the HNF4A-AS1 gene in the online genome browsers (Ensembl, UCSC, and NCBI). Which transcripts of HNF4A-AS1 have an impact on liver diseases and drug metabolism remains unclear. Some of the transcripts have different 5′ starting points due to the alternative transcription, while others differ in their exon composition due to alternative splicing. Generally, alternative transcription allows for the use of distinct promoters within a gene to initiate transcription at various starting sites, leading to multiple pre-RNA products. Different 5′ untranslated regions are considered the primary difference among these products (Pal et al., 2012; Wiesner et al., 2015), which lead to the inclusion or exclusion of certain regulatory elements, affecting RNA stability (Komarova et al., 2005; Chen et al., 2022). On the other hand, alternative splicing is a posttranscriptional process where introns are removed from the pre-RNA (Wright et al., 2022). This process can generate diverse mature RNA transcripts from a single pre-RNA by including exon skipping, intron inclusion, mutually exclusive exons, and alternative 5′ and 3′ splice sites (Gurnari et al., 2021). This dual mechanisms in producing alternative HNF4A-AS1 transcripts raise the difficulty of understanding its biological significance.

Identifying the specific functions of alternative transcripts is the best way to help us understand their significance. Alternative transcripts originating from a single pre-lncRNA can exhibit similar or different roles due to their differences in exons resulting from alternative splicing (Chen et al., 2021). It was reported that lncRNA PVT1 (full length) and its variant PVT1ΔE4 (exon 4 skipping) both promoted renal cancer cell proliferation, migration, and invasion (Yang et al., 2017). However, not all the alternative transcripts from the same pre-lncRNA show the same functions. For example, lncRNA-PXN-AS1 has two transcripts: PXN-AS1-L (with exon 4) and PXN-AS1-S (without exon 4). PXN-AS1-L promotes tumorigenesis, while PXN-AS1-S inhibits it (Yuan et al., 2017). LINC00477 is another example. Transcript 1 of LINC00477 is a tumor suppressor that is downregulated in gastric cancer tissues, while upregulated transcript 2 of LINC00477 does not significantly affect cancer cell proliferation and migration (Zhao et al., 2019). Alternative transcripts of lncRNA HNF1A-AS1 have impacts on hepatocyte cell growth, differentiation, and liver diseases and in response to drug induction (Jin et al., 2024). All these reports indicate that it is not rigorous to describe the function of all transcripts from a single lncRNA at an overall level, as different transcripts may have distinct functions.

Given the demonstrated diversity in function among alternative transcripts of lncRNAs, our attention now turns specifically to HNF4A-AS1. The HNF4A-AS1 gene is located on the genomic antisense strand of human chromosome 20 from 44,372,746 to 44,395,706. HNF4A-AS1 is transcribed antisense to the HNF4A gene. Notably, HNF4A-AS1 has a total of four annotated transcripts in the Ensembl, UCSC, and NCBI genome browsers. Dissection of the specific characteristics and roles of each transcript becomes essential for understanding the roles of HNF4A-AS1 transcripts in liver diseases and drug metabolism.

Materials and Methods

Chemicals and Reagents

HepaRG cells were kindly provided by Biopredic International (Rennes, France). HepG2 was purchased from ATCC (Manassas, VA). HepaRG growth additives (catalog number: ADD710), and HepaRG differentiation additives (catalog number: ADD720) were purchased from Biopredic International (Rennes, France). Dulbecco’s modified Eagle’s medium, fetal bovine serum, and PenStrep were obtained from GIBCO (Grand Island, NY). Williams’ E medium, Glutamax supplement, dNTPs, DreamTaq DNA polymerase, 100 bp ladder were obtained from Thermo Fisher Scientific (Carlsbad, CA). The TRIzol reagent was obtained from Invitrogen (Carlsbad, CA) and iScript cDNA Synthesis Kit, iTaq Universal SYBR Green Supermix from Bio-Rad Laboratories (Hercules, CA). The agarose gels were from IBI Scientific (Dubuque, IA), ethidium bromide from Sigma-Aldrich (Burlington, MA), and QIAquick gel extraction kit from QIAGEN (Germantown, MD).

Human Liver Samples Collection

Liver RNA samples from healthy individuals (n = 4), metabolic dysfunction-associated steatotic liver disease (MASLD) patients (n = 5), and patients with alcohol-associate liver disease (ALD) cirrhosis (n = 5) were provided by the University of Kansas Liver Tissue Biorepository, funded by the National Institute of General Medical Sciences grant 1P20GM144269-01. Acquisition of these samples was in strict adherence to ethical standards and with properly obtained informed consent.

In addition, liver RNA samples from healthy individuals (n = 4) and obese patients (n = 4) were sourced from the Department of Pharmacology and Toxicology at Rutgers University. These biopsies of obese patients (body mass index > 35, 18 < age < 80) were extracted from the left liver lobe using wedge or needle core methods. A portion of these samples was used for standard pathological evaluation, while the rest was used for research purposes. Control samples (body mass index < 30) were procured through established networks, namely the Cooperative Human Tissue Network and the National Disease Research Interchange (https://www.chtn.org/about/index.html and https://ndriresource.org/about-us). From both organizations, a comprehensive dataset including demographic information (age, race, and gender), tissue diagnosis, and quality control were collected, alongside deidentified pathology reports adhering to confidentiality protocols. The Rutgers Biomedical Health Sciences Institutional Review Board determined the use of these anonymized specimens as exempt from requiring consent under the protocols Pro2019001020 and Pro2020002744. The ethical conduct of this research was upheld through approvals by the Rutgers Biomedical Health Sciences Institutional Review Board, ensuring compliance with ethical standards and the protection of participant welfare under the specified protocols.

Differentiation of Human Embryonic Stem Cells to Hepatocytes In Vitro

Embryonic stem cells were cultivated on Matrigel with an mTeSR1 medium at a density of 35 μg/cm until the colonies reached 1 to 2 mm in diameter. To begin differentiation, a method similar to that described in previous studies was employed (Cai et al., 2007; Krueger et al., 2013). Specifically, on the first day, mTeSR1 was replaced with an induction medium designed to foster the formation of definitive endoderm. This induction medium consisted of RPMI 1640 supplemented with 0.3% bovine serum albumin, 1 × nonessential amino acids (Invitrogen Inc.), 2 mM glutamine (Invitrogen Inc.), and 100 ng/ml Activin A (Prospec Inc.). On the second day, the induction medium was refreshed, including 0.1 × insulin transferrin selenium complexes (Sigma). On the third day, the same medium was replaced again but containing 1 × insulin transferrin selenium. The generation of hepatocellular lineages commenced on the fourth day, utilizing hepatocyte culture medium (HCM) enriched with 20 ng/ml BMP4 (Prospec Inc.) and 10 ng/ml FGF2 (R&D Systems) for 5 days. By the ninth day, hepatoblasts were developed, which were then subjected to HCM supplemented with 20 ng/ml hepatocyte growth factor (Peprotech Inc.) for another 5 days to promote the formation of hepatocyte-like cells. Subsequently, these cells were further cultured in HCM with 10 ng/ml oncostatin M and 0.1 μM dexamethasone for an additional 5 days to achieve maturation into hepatocyte-like cells.

Cell Culture

HepG2 cells were cultured in Dulbecco’s modified Eagle’s medium with 10% fetal bovine serum based on the provider’s instruction. HepaRG cells were cultured according to the provider’s protocol. Briefly, the HepaRG cells were primarily cultured in a HepaRG growth medium (Williams’ E medium supplied with Glutamax and growth additives) for 14 days until cells became fully confluent. Cells were then kept in a HepaRG differentiation medium for 2 more weeks to make sure they were fully differentiated.

During the experiment time, HepG2 and HepaRG cells were incubated at 37°C and 5% CO₂, and the medium was renewed every 3 days. Then, HepG2 cells and fully differentiated HepaRG cells were used directly in the T-25 flask or seeded into 12-well plates for further experiments.

RNA Isolation and cDNA Synthesis in Cells and Human Samples

Total RNA was extracted from ∼50 mg human liver samples using the miRNeasy column purification kit (Qiagen 217004) with on-column DNase treatment using the RNase-Free DNase Set (Qiagen 79254) according to the manufacturer’s instructions. Nuclease-free water was used for the final elution of RNA from the purification column. Concentrations and O.D. 260/280 purity ratio were determined by a Nanodrop 8000 (Thermo Fisher).

Total RNA was extracted from HepG2 or HepaRG cells using a TRIzol reagent following the manufacturer's instructions. RNA concentrations and purity were quantified at 260 nm with a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). cDNA was synthesized from 1 μg of total RNAs using an iScript cDNA Synthesis Kit.

Polymerase Chain Reaction

To visualize the alternative transcripts of HNF4A-AS1, polymerase chain reaction (PCR) was initially performed to amplify each transcript from the cDNA extracted from HepG2, HepaRG cells, and human liver samples. Four specific pairs of designed primers targeting four alternative transcripts were used in PCR to amplify the target regions (Table 2). The PCR reaction mixture comprised the cDNA template, primers, deoxynucleotide triphosphates, and DreamTaq DNA polymerase.

TABLE 2.

Information on PCR primers for amplifying each annotated transcript of HNF4A-AS1

Primer Name	Forward Primer (5′-3′)	Reverse Primer (5′-3′)	Product Length (bp)
E16	CTGGTCTTGCTGCTTCCT	AGCCTGGAATAGCAGCATC	655
E15	CTGGTCTTGCTGCTTCCT	GGGGCTAATAGGGTAGTGG	457
E14	CTGGTCTTGCTGCTTCCT	AATATGACCGGTGTGCAGTC	283
E35	GTCACACCTGGGCAGAAG	GGTGTCATGGCTTCTCTGG	344

Gel Electrophoresis

The PCR product from each primer pair was detected by electrophoresis to validate the presence of alternative transcripts of HNF4A-AS1. Products were loaded into 2% agarose gels and run under a constant voltage of 120 V for around 30 minutes. After that, gels were stained with ethidium bromide solution before being visualized under ChemiDoc MP.

PCR Product Purification and Sanger Sequencing

PCR products were purified from the gel using the QIAquick Gel Extraction Kit following the manufacturer’s instructions. The purified DNA samples were then sent to Eurofins Scientific (Boston, MA) for Sanger sequencing to identify the nucleotide sequence of the amplified fragments.

Quantitative PCR

Using the cDNA from each group in this study as a template, quantitative PCR (qPCR) was conducted with iTaq Universal SYBR Green Supermix reagent according to the manufacturer's instructions. The annealing temperature was set at 60°C. β-actin was used as the internal control for qPCR. All qPCR primers are listed in Table 3.

TABLE 3.

Information on qPCR primers

Primer Name	Forward Primer (5′-3′)	Reverse Primer (5′-3′)	Product Length (bp)
E16	AAGCTTTAAAGGCCAGTGC	TGCTTCACTTCTCTGGTCC	177
E16B2	GGCAGATACCAGCTCATGG	CTTGAGGACAGATCAGAGGC	165
E15	CCAACTCCACCTCCCAAATTAC	CCAGAACTCCAGGCATGTCT	197
E14	CTCTTCCTGGAGCTGGGATC	AATATGACCGGTGTGCAGTC	148
E35	AAGCTTTAAAGGCCAGTGC	TTGGGTTAGAGCTACATCCAC	177
CYP3A4	GATTGACTCTCAGAATTCAAAAGAAACTGA	GGTGAGTGGCCAGTTCATACATAATG	148
β-ACTIN	GGACTTCGAGCAAGAGATGG	AGCACTGTGTTGGCGTACAG	234

Data Analysis

The data are presented as means ± S.D. Statistical analyses for data were performed using GraphPad Prism (La Jolla, CA). Statistically significant differences in gene expression between groups were determined by using the one-way ANOVA test followed by Tukey's post hoc test. Significant P-values are indicated with asterisks as follows: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Pearson correlation was used for the correlation analyses.

Results

Validation of Alternative Transcripts of HNF4A-AS1 in Liver Cell Lines and Human Liver

HNF4A-AS1 is a lncRNA located in human chromosome 20 from 44,372,746 to 44,395,706 with a total of six exons in between (Fig. 1A). It has been annotated to contain multiple alternative transcripts from various genome browsers (NCBI, Ensemble, and UCSC). However, inconsistencies in these alternative transcripts exist among these databases. Specifically, NCBI annotates three alternative transcripts—NR_172878.1, NR_172879.1, and NR_109949.1—while Ensemble and UCSC give two alternative transcripts: ENST00000452481.1 and ENST00000609152.1. After integrating their information, HNF4A-AS1 has a total of four alternative transcripts (Table 1). Based on the first exon and last exon that are contained in each transcript, these transcripts were given shorter names: E16, E15, E14, and E35 (Table 1). It was notable that all information regarding these alternative transcripts was based solely on RNA sequencing annotation and had not been validated using in vitro or in vivo models. Therefore, it was crucial to determine which alternative transcripts truly exist in the human liver and different human liver cell lines.

Fig. 1.

Validation of alternative transcripts of HNF4A-AS1 in liver cell lines and human liver. (A) The lncRNA gene HNF4A-AS1 is on chromosome 20 with six exons in between. (B) The schematic diagram of four confirmed annotated HNF4A-AS1 transcripts (E16, E15, E14, and E35) and two novel identified transcripts (E16B2 and E35B2) are shown. The locations of PCR primers are depicted by a black arrow (forward primer) and a white arrow (reverse primer). (C) In HepaRG cells, four annotated transcripts (E16, E15, E14, and E35) and two new transcripts (E16B2, E35B2) were amplified. There was one mismatched band in the top of the E16 lane (>600 bp). (D) In HepG2 cells, four annotated transcripts (E16, E15, E14, and E35) and two new transcripts (E16B2 and E35B2) were identified. There were four mismatched bands (all of them > 600 bp). (E–H) In four different individual normal patients (body mass index < 30), only two annotated transcripts (E16 and E14) and one new transcript (E16B2) were confirmed. The sequences of all annotated and newly identified transcripts are listed in Supplemental Table 1.

TABLE 1.

Information of HNF4A-AS1 alternative transcripts from NCBI, UCSC, ensemble genome browsers, and experimental confirmation

Transcript ID	Length (nt)	First Exon	Last Exon	Total Exons	Abbreviation
NR_172878.1	4538	1	6	3	E16
NR_172879.1	1242	1	5	3	E15
NR_109949.1/ ENST00000452481.1	648	1	4	4	E14
ENST00000609152.1	413	3	5	2	E35
N/A	4388	1	6	2	E16B2
N/A	506	3	5	3	E35B2

To validate the existence of alternative transcripts of HNF4A-AS1, we designed specific four pairs of PCR primers as demonstrated in Fig. 1B and listed in Table 2 to amplify each transcript on HepaRG cells, HepG2 cells, and human liver samples via PCR. The gel results showed that seven bands existed in HepaRG cells, including three bands in E16, one band in E15, one band in E14, and two bands in E35 (Fig. 1C). We purified DNA from each band for Sanger sequencing and confirmed, among the seven bands in HepaRG cells, one band was a mismatch and four of them matched the HNF4A-AS1 transcripts in the genome browser while the other two were novel HNF4A-AS1 transcripts. In E16, the upper light band was a mismatch; the bright middle band matched transcript NR_172878.1, so it kept the name E16, while the lower band had an exon 3 skipping compared with E16; as a result, it had a new name, E16B2. Similarly, in E35, the brighter lower band matched transcript ENST00000609152.1, so it kept the name E35. The upper band was very light and had an intron retention compared with E35; thus it had a new name, E35B2. The band in E15 matched NR_172879.1 (E15) and the band in E14 matched NR_109949.1/ENST00000452481.1 (E14) (Fig. 1C). When it came to HepG2, there were more unspecific bands (over 600 bp) that existed in E16 and E15 (Fig. 1D). Except for those unspecific bands, Sanger sequencing results showed that HepG2 had the same bands as those in HepaRG cells. To discover how many transcripts exist in vivo, the same experiments were conducted on human liver samples from four different patients. E16, E16B2, and E14 existed in all human samples, while E15, E35, and E35B2 were not detected in any samples (Fig. 1, E–H). Interestingly, the relative expression levels of E16 and E16B2 varied among different human samples. In samples 1 and 3, the expression levels of E16 and E16B2 were relatively similar (Fig. 1, E and G), while in samples 2 and 4, E16 had an obviously higher expression level than E16B2 (Fig. 1, F and H). All sequence information of both annotated and newly identified alternative transcripts is listed in Supplemental Table 1. All gel results revealed that the different transcripts of HNF4A-AS1 not only showed in vitro and in vivo differences but also existed the interindividual differences.

Expression Profiles of HNF4A-AS1 Alternative Transcripts under Different Human Liver Disease Conditions

After validating the existence of alternative transcripts of HNF4A-AS1 in liver cell lines and human liver samples, we next examined their expression profiles under various human liver disease conditions. Normal liver samples (n = 8), MASLD samples (n = 5), ALD cirrhosis samples (n = 5), and liver samples from obese patients (n = 4) were collected to assess the relative RNA levels of HNF4A-AS1 via reverse transcription (RT)-qPCR method. Unfortunately, it was unable to design a specific pair of primers to detect transcript E35B2 by qPCR due to its sequence limitation. Thus, only five out of six alternative transcripts of HNF4A-AS1 were studied for their expression patterns with corresponding qPCR primers for each transcript listed in Table 3. The results showed that transcript E16 increased significantly in the ALD cirrhosis samples compared with the normal samples, indicating E16 may have a role in ALD cirrhosis disease (Fig. 2A). Even though the results in other groups did not have statistical differences, some changing trends of expression levels still can be observed. E16B2 exhibited increased trends in patients with MASLD and ALD cirrhosis compared with the expression levels in normal patients (Fig. 2B). E15 displayed an increased trend only in MASLD patients (Fig. 2C). E14 and E35 showed no changes in trends among normal, MASLD, ALD cirrhosis, and obese patients (Fig. 2, D and E). These results suggest potential roles of some HNF4A-AS1 transcripts (E16B2 and E15) in MASLD or ALD cirrhosis diseases. The small sample sizes in this study may contribute to the insignificant difference, and more samples should be included in future studies to gain a more accurate expression pattern for each transcript of HNF4A-AS1.

Fig. 2.

Expression profiles of HNF4A-AS1 alternative transcripts under different human health conditions. RT-qPCR was applied to quantify the expression levels of HNF4A-AS1 different transcripts, including E16 (A), E16B2 (B), E15 (C), E14 (D), and E35 (E) in four different conditions: normal (n = 8), MASLD (n = 5), ALD cirrhosis (n = 5), and obesity (body mass index > 35) (n = 4). The data are presented as mean ± S.D. (n = 3).

Expression Profiles of HNF4A-AS1 Alternative Transcripts during Induction of Embryonic Stem Cells to Matured Hepatocyte-like Cells

Understanding the characteristics of different alternative transcripts of lncRNA HNF4A-AS1 during hepatocyte maturation is important. In this study, human embryonic stem cells were initially inducted into definitive endoderm in vitro, followed by hepatoblast cells, then differentiated into hepatocyte-like cells (HLCs), and finally formed matured HLCs. The whole process was reported to mimic the in vivo hepatocyte maturation (Cai et al., 2007; Krueger et al., 2013). RT-qPCR was applied to examine how these alternative transcripts expressed in different stages of hepatocyte maturation. As shown in the results, all transcripts significantly increased their expression at the hepatoblast stage. E16 and E16B2 kept high expression levels at both hepatoblasts and HLC stages then decreased at matured HLC stages (Fig. 3, A and B). E15, E14, and E35 also significantly increased their expression at hepatoblasts stages but gradually decreased in HLC and matured HLC stages (Fig. 3, C–E). However, different transcripts of HNF4A-AS1 showed distinct relative expression levels during the formation of matured HLCs. Transcript E16 was the most abundant in all stages. E15 and E35 also had relatively high expression levels, while E16B2 and E14 expression levels were low (Fig. 3F). These results suggest that these transcripts may have functions during the formation of hepatocytes.

Fig. 3.

Expression profiles of HNF4A-AS1 alternative transcripts during induction of hESCs to matured HLCs. RT-qPCR was applied to quantify the expression levels of HNF4A-AS1 alternative transcripts, including E16 (A), E16B2 (B), E15 (C), E14 (D), and E35 (E) at different stages from hESC to matured HLCs. (F) Relative expression levels of all transcripts of HNF4A-AS1 at different stages from hESC to matured HLCs are shown. Each group was compared with the group on its left. The data were presented as mean ± S.D. (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. hESC, human embryonic stem cell.

Expression Profiles of HNF4A-AS1 Alternative Transcripts during Growth and Differentiation Periods of HepaRG Cells

The HepaRG cell is an important hepatic cell line for drug metabolism related studies because it retains many characteristics of primary human hepatocytes, especially the high levels of CYP enzymes. This cell line undergoes a long period of culture with 2 weeks for growth and another 2 weeks for differentiation, which can be an indication of hepatocyte maturation. Exploring the expression patterns of different alternative transcripts of HNF4A-AS1 and their correlations with CYPs during HepaRG cells maturation is meaningful. CYP3A4 is a key CYP isoform involved in metabolizing a large portion of clinically used drugs (Evans and Relling, 1999). As a result, we used CYP3A4 as a model CYP in this study. HepaRG cells were cultured in proper conditions and harvested every 2 to 3 days until they completely differentiated. Specifically, for the growing period, cells were harvested on days 3, 6, 9, 12, and 14 (end point of growth) and later harvested on days 17, 20, 23, 26, and 28 (end point of differentiation). RT-qPCR was conducted to test different transcripts of HNF4A-AS1 and CYP3A4 at these time points. The results exhibited a consistent pattern in all transcripts of HNF4A-AS1. All transcripts showed an extremely significant increase in their expression levels on day 17, the initiation of cell differentiation, followed by a gradual decline until day 28 (Fig. 4, A–E). In addition, when comparing the relative expression levels of the alternative transcripts, it was observed that transcript E16 was the most abundant, whereas transcript E14 demonstrated limited expression levels in HepaRG cells (Fig. 4G). CYP3A4 was also measured during the growth and differentiation periods and had the same expression trends as HNF4A-AS1 transcripts (Fig. 4F). This phenotype raised an additional inquiry regarding the potential correlation between expression profiles of alternative transcripts of HNF4A-AS1 and CYP3A4. To address this question, Pearson correlation was conducted between each transcript and CYP3A4. The results demonstrated a strong positive correlation between each alternative transcript and CYP3A4 (P < 0.0001), suggesting their potential regulatory roles in CYPs (Fig. 4, H–L).

Fig. 4.

Expression profiles of HNF4A-AS1 alternative transcripts during growth and differentiation periods of HepaRG cells. RT-qPCR was applied to quantify the expression levels of HNF4A-AS1 alternative transcripts, including E16 (A), E16B2 (B), E15 (C), E14 (D), and E35 (E) as well as CYP3A4 (F) at every 2 to 3 days during HepaRG maturation from day 3 until day 28. (G) Relative expression levels of all transcripts of HNF4A-AS1 at different days of HepaRG maturation are shown. Each group was compared with the group on its left. The data were presented as mean ± S.D. (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (H–L) Pearson correlation analysis was conducted to analyze the relationship between each transcript of HNF4A-AS1, E16 (H), E16B2 (I), E15 (J), E14 (K), and E35 (L) with CYP3A4 expression levels. All the transcripts showed strong positive relationships with CYP3A4.

Expression Profiles of HNF4A-AS1 Alternative Transcripts in Response to Xenobiotic Exposure

The expression of some CYP enzymes could be induced under certain conditions, including drug treatment. Rifampicin (RIF), an antibiotic used in the treatment of gram-positive bacteria, is known as a CYP inducer, including CYP2B6, 2C8, 2C9, 2C19, and 3A4/5 (Chen and Raymond, 2006; Singh et al., 2012). To explore the roles of different alternative transcripts of HNF4A-AS1 dealing with CYPs under the RIF exposure condition, matured HepaRG was treated with 10 μM RIF for 5 days and the cells were harvested at different time points at 12, 24, 36, 48, 60, 72, 84, 96, 108, and 120 hours. Later, RT-qPCR was used to determine the relative expression levels of alternative transcripts of HNF4A-AS1. The results revealed that E16 and E16B2 exhibited similar patterns: their expression levels were relatively low at the first 48 hours and sharply rose to peak levels at 60 hours, then gradually decreased thereafter until 120 hours (Fig. 5, A and B). E15 did nott show significant change in the first 48 hours but increased at 60 hours and kept the expression level (Fig. 5C). E14 did not show significant change in the first 48 hours but started increasing from 60 hours and reached a peak at 96 hours (Fig. 5D). E35 had similar expression patterns as E16 in the first 48 hours and had similar expression patterns as E14 after 48 hours (Fig. 5E). Comparing the expression levels of all transcripts, the transcript E16 was still the most abundant, followed by transcript E15, E35, E16B2, and E14 (Fig. 5G).

Fig. 5.

Expression profiles of HNF4A-AS1 alternative transcripts in response to xenobiotic exposure. RT-qPCR was applied to quantify the expression levels of HNF4A-AS1 alternative transcripts, including E16 (A), E16B2 (B), E15 (C), E14 (D), and E35 (E) as well as CYP3A4 (F) at every 12 hours after treatment of 10 μM rifampicin from 12 hours to 120 hours. (G) Relative expression levels of all transcripts of HNF4A-AS1 at different time points after the treatment of rifampicin. Each group was compared with the group on its left. The data were presented as mean ± S.D. (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. (H–L) Pearson correlation analysis was conducted to analyze the relationship between each transcript of HNF4A-AS1 with CYP3A4 expression levels. E16 (H), E16B2 (I), and E15 (J) showed strong positive relationships with CYP3A4 while E14 (K) and E35 (L) did not have relationships with CYP3A4.

As RIF could induce the expression of CYP3A4, CYP3A4 showed no change at first 48 hours, sharply reached a peak expression level at 60 hours, and maintained the platform until 84 hours, followed by decline (Fig. 5F). Pearson correlations were used to analyze the relationships between each alternative transcript of HNF4A-AS1 and CYP3A4. Interestingly, E16, E16B2, and E15 all had a strong positive correlation with CYP3A4 (P < 0.0001) (Fig. 5, H–J), while the other two transcripts (E14 and E35) did not have statistical significance in the correlation with CYP3A4 (Fig. 5, K and L). These results strongly suggest that some of the alternative transcripts of lncRNA HNF4A-AS1 (E16, E16B2, E15) might play roles on CYP expression under the induction by RIF.

Discussion

This study advances our understanding of HNF4A-AS1 and its alternative transcripts by identifying and characterizing these transcripts, examining their expression across various conditions, and analyzing their correlation with CYP3A4. These insights pave the way for a better understanding the role of HNF4A-AS1 in liver diseases and drug metabolism.

The transcript profiles of HNF4A-AS1 in liver tissues can differ significantly from those in established cell lines. Integrating the information of HNF4A-AS1 transcripts in genome browsers, there are four annotated transcripts (E16, E15, E14, and E35). Our data showed that there was no difference in the transcript profiles of HNF4A-AS1 between the HepaRG and HepG2 cell lines. They both contained the four annotated transcripts along with two newly identified transcripts (E16B2 and E35B2) of HNF4A-AS1. This similarity is likely because both cell lines originate from hepatocellular carcinomas, which might share certain transcriptional characteristics (Aden et al., 1979; Hart et al., 2010). However, the results in human liver tissues showed a big difference from those in cell lines. In liver samples from four individuals, only E16, E16B2, and E14 were present. This variation suggests that tumor-derived cell lines might carry genetic and epigenetic alterations that do not precisely reflect the normal liver tissues (Domcke et al., 2013). Another notable discovery is the variable relative expression levels of HNF4A-AS1 transcripts among individuals, especially between E16 and E16B2. In liver samples 1 and 3, the expression levels of E16 and E16B2 were similar, whereas in liver samples 2 and 4, E16 had more expression levels than E16B2. This interindividual variability indicates the complexity of studying alternative transcripts and highlights the need for precise characterization in future research on lncRNA functions and their regulatory mechanisms.

The expression of HNF4A-AS1 transcripts in liver diseases provides clues for targeted RNA therapeutics. The increase in the expression of E16 in ALD cirrhosis compared with normal liver samples suggests that E16 may be a critical RNA molecule for targeting ALD cirrhosis. However, we acknowledge the limitation of our current sample size, which may affect the validation of our conclusions. Unfortunately, existing data from publicly available databases, such as Gene Expression Omnibus, do not currently provide detailed information on different transcripts of HNF4A-AS1 across various liver diseases, making it challenging to validate our findings using external datasets. Therefore, while our study offers valuable initial insights, future research should involve larger and more diverse cohorts to confirm the expressions of these transcripts in liver diseases.

In addition, our study indicates that HNF4A-AS1 transcripts play a role in hepatocyte formation. In vitro differentiation of human embryonic stem cells into matured HLCs serves as a parallel to in vivo hepatocyte maturation (Cai et al., 2007; Krueger et al., 2013). In this study, a noticeable increase in the expression of all HNF4A-AS1 transcripts was observed during the hepatoblast stage, a critical phase where cells are hepatic progenitors (Li et al., 2010), indicating potential involvement of HNF4A-AS1 transcripts in early liver development. HepaRG is a hepatic progenitor cell line with bipotential that can differentiate into hepatocytes or cholangiocytes (Hart et al., 2010; Andersson et al., 2012). The standard differentiation protocol for HepaRG cells involves an initial 14-day growth phase, during which the cells proliferate, followed by a 14-day differentiation phase. During the latter phase, the cells mature into a mixed population of hepatocyte-like and cholangiocyte-like cells. This 28-day development period effectively models the stages of hepatic growth and differentiation, providing a reliable in vitro system to study liver development and function. Remarkably, all the HNF4A-AS1 transcripts maintained a low expression level during the initial growth phase but experienced a significant increase during the differentiation phase, showing the crucial role in the transition from progenitor to fully differentiated hepatocytes. All these findings reinforce the hypothesis that HNF4A-AS1 plays a role in hepatocyte formation, offering targets for RNA-based therapies in liver regeneration and development.

In addition, several HNF4A-AS1 transcripts are involved in drug-drug interactions (DDIs) which are mainly caused by the induction or inhibition of drug-metabolizing enzymes (Hakkola et al., 2020). RIF, known as the potent CYP inducer, frequently causes DDIs by enhancing the activity of these enzymes (Chen and Raymond, 2006). Among them, CYP3A4 is one of the major responsive CYPs (Kanebratt et al., 2008; Hamilton et al., 2014). The dynamic expression patterns of CYP3A4 following RIF administration in HepaRG cells offers insights into cellular mechanisms for xenobiotic handling. CYP3A4 exhibited a peak response window from 60 to 84 hours posttreatment. Interestingly, only E16, E16B2, and E15 showed similar expression patterns and had strong positive relationships with CYP3A4. This observation implies that these specific transcripts might be involved in HepaRG cells responding and metabolizing of RIF treatment, indicating the potential of RNA-based therapies to modulate specific transcripts to manage DDIs effectively.

Understanding the specific functions of alternative transcripts is crucial for RNA therapeutic strategies. Alternative transcripts of a single lncRNA may share the same functions if they retain essential functional sites after alternative splicing. Conversely, their functions might differ if key functional domains are removed through splicing (Chen et al., 2021). For example, both the full-length lncRNA PVT1 and its variant PVT1ΔE4 (without exon 4) promoted the progression of renal cancer because the critical functional sites were not located on exon 4 (Yang et al., 2017). Conversely, PXN-AS1-L (with exon 4) promoted tumorigenesis, whereas PXN-AS1-S (without exon 4) prevented it, suggesting the functional significance of exon 4 (Yuan et al., 2017). Since some transcripts of HNF4A-AS1 share the same exons (such as E16, E16B2, E15, and E14 all including exon 1) and some transcripts of HNF4A-AS1 do not have any overlap with each other (like E16 and E35), there is potential for these transcripts to exhibit either similar or distinct functions.

Future research should focus on specifying targeted transcripts to improve RNA-based therapeutic strategies. In the papers related to HNF4A-AS1, some papers did not specify which transcript was under investigation. Researchers often used commercially available small interfering RNAs (siRNAs) or small hairpin RNAs without considering which transcripts they were targeting. For instance, a commonly used siRNA for HNF4A-AS1 (siRNA ID: n356309) aimed at exon 3, impacted E16, E15, E14, E35, and E35B2 simultaneously. However, the efficiency of the knockdown was assessed using a TaqMan probe (Hs01378672_m1) that only targeted E14 (Chen et al., 2018). Additionally, some studies did not specify the exact locations targeted by small hairpin RNAs (Chen et al., 2020; Wang et al., 2021), which could lead to misunderstandings. This oversight might cause confusion and inaccuracies in interpreting the effects of gene silencing. Future studies should emphasize the specific RNA ID to clearly indicate the transcript they are investigating. As a result, precision medicine also can gain better development based on this improvement.

The present study has provided important insights into the expression patterns of HNF4A-AS1 transcripts in liver cells and tissues, particularly in the context of drug metabolism. However, several critical questions remain unanswered. The distinct or overlapping functions of the different HNF4A-AS1 transcripts are still not fully understood, and future studies should focus on elucidating whether these transcripts act redundantly or have unique roles in liver function and disease processes. Additionally, although the involvement of HNF4A-AS1 transcripts in hepatocyte differentiation has been observed, the precise mechanisms through which these transcripts influence this process require further investigation. Overexpression and knockdown experiments targeting specific transcripts, particularly HNF4A-AS1-E16, could shed light on their regulatory effects on key metabolic enzymes like CYP3A4, given the observed correlation between HNF4A-AS1-E16 expression and CYP3A4 levels in liver models. Moreover, it is crucial to determine whether the regulatory effects of HNF4A-AS1-E16 on drug metabolism and liver disease are mediated dependent or independent of the HNF4A gene. Understanding this relationship will be vital for the development of RNA-based therapies targeting HNF4A-AS1 without disrupting the broader regulatory network governed by HNF4A. These future research directions are essential for building on the foundation laid by our current findings and translating them into practical applications in liver diseases and drug metabolism.

In summary, this study investigated the unique expression patterns of different transcripts of HNF4A-AS1 under various biological scenarios. This knowledge is crucial for developing targeted RNA-based therapeutic strategies, particularly in the context of liver diseases and drug metabolism. By focusing on the specific roles of each transcript, we can significantly advance precision medicine, leading to more effective and personalized treatments.