LncRNA LL22NC03-N14H11.1 promoted hepatocellular carcinoma progression through activating MAPK pathway to induce mitochondrial fission

Involvement of long non-coding RNAs (lncRNAs) in hepatocarcinogenesis has been largely documented. Mitochondrial dynamics is identified to impact survival and metastasis in tumors including hepatocellular carcinoma (HCC), but the underlying mechanism remains poorly understood. This study planned to explore the regulation of lncRNA LL22NC03-N14H11.1 on HCC progression and mitochondrial fission. Dysregulated lncRNAs in HCC are identified through circlncRNAnet and GEPIA bioinformatics tools. Biological function of LL22NC03-N14H11.1 in HCC was detected by CCK-8 assay, flow cytometry analysis, transwell invasion, and wound healing assays. Molecular interactions were determined by RNA immunoprecipitation, RNA pull-down, and co-immunoprecipitation assays. Results showed that LL22NC03-N14H11.1 was upregulated in HCC tissues and cells. Functionally, LL22NC03-N14H11.1 contributed to cell proliferation, migration, invasion, and epithelial-to-mesenchymal transition (EMT) in HCC. Moreover, LL22NC03-N14H11.1 facilitated mitochondrial fission in HCC cells. Mechanistically, LL22NC03-N14H11.1 recruited Myb proto-oncogene (c-Myb) to repress the transcription of leucine zipper-like transcription regulator 1 (LZTR1), so as to inhibit LZTR1-mediated ubiquitination of H-RAS (G12V), leading to the activation of mitogen-activated protein kinase (MAPK) signaling and induction of p-DRP1 (Serine 616). In conclusion, this study firstly revealed that lncRNA LL22NC03-N14H11.1 promoted HCC progression through activating H-RAS/MAPK pathway to induce mitochondrial fission, indicating LL22NC03-N14H11.1 as a novel potential biomarker for HCC treatment.


Introduction
Hepatocellular carcinoma (HCC) is the third most fatal malignancy globally, taking up 90% of primary liver cancer cases 1 . Although during recent decades, surgery and interventional therapies have been greatly advanced, survival of HCC patients is still unsatisfactory 2 , which is mainly attributed to easy metastasis and frequent recurrence 3 . Therefore, further exploration of the precise mechanism behind HCC progression is urgently required.
Long non-coding RNAs (lncRNAs) are defined as a group of transcripts with a length over 200 nucleotides and lack of protein products 4,5 . Volumes of studies have offered convincing evidences that lncRNAs serve important roles in carcinogenesis of a wide range of cancer types 6,7 , including HCC 8,9 . Mechanistically, lncRNAs functioned through regulating target genes at epigenetic, transcriptional, and post-transcriptional levels 10,11 , and the mechanism varies depending on the cellular location of lncRNAs 12 . Here, we firstly discovered a novel lncRNA LL22NC03-N14H11.1 located at chr22: 16,154,154,766 via bioinformatics analysis, and the results showed that LL22NC03-N14H11.1 was upregulated and had poor prognostic significance in HCC samples. Thus, we speculated that LL22NC03-N14H11.1 might participate in HCC progression and intended to probe into its role in HCC.
The morphology of mitochondria is highly dynamic and varies constantly through fission and fusion, helping cells adjust their morphology to satisfy the variable need of cells and adapt to the cellular environment 13 . Studies have shown that mitochondrial fission is increased in tumor cells, and that inhibiting mitochondrial fission can impede proliferation and induce apoptosis in several cancer types, such as colon cancer and lung cancer [14][15][16] . Also, facilitated mitochondrial fission can aggravate cancer metastasis through promoting migration and invasion 17,18 . Notably, Huang et al. 19 stated that mitochondrial fission was increased in HCC and promoted HCC cell survival and autophagy. However, whether LL22NC03-N14H11.1 played a part in HCC via its regulation on mitochondrial fission has never been explored.
Moreover, highly conserved dynamin-related GTPases are recognized as primary regulators of mitochondrial dynamics, and the typical one of which is dynamin-related protein 1 (DRP1) 13,20 . Phosphorylation of DRP1 at Serine 616 (S616) is proved to be required for the activation and recruitment of DRP1 to the mitochondrial membrane to induce mitochondrial fission 21 . For example, CDK1 phosphorylates Ser585 of rat DRP1, which is equivalent to human S616 during mitosis 22 . Protein disulfide isomerase (PDI1) facilitates DRP1 S616 phosphorylation and mitochondrial fission in CA1 neurons 23 . Multiple studies reveal that active ERK1/2 also contributes to the phosphorylation of DRP1 at S616 24 . Importantly, DRP1 S616 phosphorylation by the activation of RAS/MAPK (mitogen-activated protein kinase) signaling is demonstrated to be oncogenic in cancer cells. For example, RAS (G12V)induced MAPK pathway led to DRP1 S616 phosphorylation and mitochondrial fission and tumor cell survival 25 . H-RAS (G12V) phosphorylated ERK1/2 and contributed to pancreatic cancer tumor growth through facilitating DRP1 S616 phosphorylation and mitochondrial fission 26 . To date, RAS/MAPK pathway has been largely reported to be involved in cell survival, apoptosis, and metastasis in cancers. However, the correlation of LL22NC03-N14H11.1 with RAS/MAPK signaling has never been investigated yet.
Leucine zipper-like transcription regulator 1 (LZTR1) is a Golgi protein belonging to BTB-Kelch superfamily 27 , which is generally known to function through interacting with Cullin3 (CUL3)-based E3 ubiquitin ligases 28,29 . LZTR1 has been reported to potentially participate in apoptosis and ubiquitination 27,30,31 . Several studies have revealed the tumor-suppressive role of LZTR1, and the germline and somatic mutations of LZTR1 in patients with glioblastoma and schwannomatosis [32][33][34] . Interestingly, it has been reported that LZTR1 is a conserved regulator of the ubiquitination of RAS family, leading to the degradation of both wild-type and mutant RAS members to inactivate MAPK pathway 35 . These findings indicated that LZTR1 might elicit its tumor-suppressive function through MAPK pathway, and that LZTR1 might be related to mitochondrial fission. However, the function and mechanism of LZTR1 in HCC and its association with mitochondrial fission have never been revealed.
Therefore, the present study aimed to uncover the role and mechanism of LL22NC03-N14H11.1 in HCC and its potential regulation on mitochondrial fission.

Clinical specimen collection
Sixty-two patients diagnosed with HCC were recruited for collecting tumor and adjacent non-tumor tissue samples between 2013 and 2018, with the approval of Ethics Committee of Affiliated Hospital of YouJiang Medical University For Nationalities. Paired non-tumor tissues were at least 5 cm away from the tumor margin. Samples were frozen immediately in liquid nitrogen after surgical resection and stored at −80°C until used for total RNA extraction. Patients treated with radiotherapy or chemotherapy before surgery were excluded, and all subjects had provided the written informed consent.

Subcellular fractionation
Nuclear and cytoplasmic fractions of SK-HEP-1 or Huh7 cells were separated utilizing the PARIS Kit (Thermo Fisher). Cell samples were first lysed in cell fractionation buffer. The pellet washed with TSE buffer (10 mM Tris, 300 mMsucrose, 1 mM EDTA, 0.1% NP40 PH 7.5) at 4000 g for 5 min in a tabletop centrifuge at 4 ℃. The resulting supernatant discarded and the pellets were nucleus. The resulting supernatant from the first round of differential centrifugation was sedimented for 150 min at 14000 rpm in a tabletop centrifuge. The resulting pellets were membranes and the supernatant were cytoplasm. RT-qPCR was followed to determine the location of LL22NC03-N14H11.1, with GAPDH or U6 as the cytoplasmic or nuclear control. The experiment was conducted in triplicate.

Fluorescence in situ hybridization
The RNA fluorescence in situ hybridization (FISH) probe for LL22NC03-N14H11.1 was designed and produced at RiboBio (Guangzhou, China). Cell samples were fixed in 4% formaldehyde for 15 min before washing in phosphate-buffered saline (PBS), and then cultured with pepsin and dehydrated. The air-dried cells were incubated with FISH probe in the hybridization buffer. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI) and cells were observed using laser scanning confocal microscope (Zeiss, Jena, Germany). The experiment was conducted in triplicate.

CCK-8
Transfected SK-HEP-1 or Huh7 cells were prepared in 96-well plates containing complete medium with 5 × 10 3 cells in each well, and 10 µL of CCK-8 solution (Dojindo, Kumamoto, Japan) was added into each well for 2 h to detect cell viability. The optical density value at 450 nm was recorded at 0, 24, 48, 72, and 96 h using a microplate reader (Dynex Technologies, West Sussex, UK). The experiment was conducted in triplicate.

Colony formation
Transfected SK-HEP-1 or Huh7 cells at the logarithmic growth phase were trypsinized, harvested, and planted into 6-well plates for 2 weeks, at the density of 500 cells per well. After washing in PBS, colonies containing more than 50 cells were fixed in methanol (Sigma-Aldrich), dyed in crystal violet (Sigma-Aldrich), and eventually counted manually. The experiment was conducted in triplicate.

Flow cytometry of apoptosis
Annexin-V-FITC/Propidium Iodide (PI) Apoptosis Detection Kit (BD Biosciences, Franklin Lakes, NJ, USA) was obtained for flow cytometry analysis. A total of 1 × 10 6 cells of SK-HEP-1 or Huh7 were collected after transfection, and plated into 6-well plates for treatment in 100 μL of 1× Binding Buffer containing 5 μL of PI and 5 μL of Annexin-V-FITC. After culturing in a dark room for 15 min, cell apoptosis of transfected SK-HEP-1 or Huh7 cells was analyzed by FACScan (BD Biosciences) and FlowJo V10 software (Tree Star, Ashland, OR, USA). The experiment was conducted in triplicate.

Western blotting
Total cellular protein samples were prepared using RIPA lysis buffer and subjected to sodium dodecyl sulfate (SDS)-polyacrylamide gel, following moving onto PVDF membranes (Bio-Rad Laboratories, Hercules, CA, USA). After blocking with 5% nonfat milk, membranes were cultivated with primary antibodies at 4°C all night. Following washing in Tris-buffered saline/Tween-20 (TBST), membranes were probed with secondary antibodies at room temperature for 2 h. The band density was analyzed with application of the ECL luminous liquid (Pierce, Rockford, IL, USA) and quantified via the ImageJ software (National Institutes of Health, USA). The experiment was conducted in triplicate.

Invasion assay
Cell invasion was explored by 24-well Transwell chambers (8-mm pore size; Corning Incorporated, Big Flats, NY, USA) pre-coated with Matrigel (BD Biosciences). The upper chamber was added with 2 × 10 4 transfected cells in serum-free medium, while lower chamber was filled with complete culture medium containing 10% FBS. Invaded cells on the bottom of membrane were fixed by 4% formaldehyde after 24 h, stained with crystal violet, and counted under the optical microscope (Thermo Fisher). The experiment was conducted in triplicate.

Wound healing
Transfected SK-HEP-1 or Huh7 cells (1 × 10 6 ) were seeded into 6-well plates and cultured until cells reached 80-90% confluence. The following day, cells were treated with a 200 μL filter tip for creating the wound area, followed by 24 h of incubation at 37°C in serum-free medium. The distance of wound healing was photographed under a microscope (Olympus, Tokyo, Japan) at 0 or 24 h. The experiment was conducted in triplicate.

Immunofluorescence
Transfected SK-HEP-1 or Huh7 cells in PBS (Sigma-Aldrich) were plated onto culture slides for 24 h, incubated in 1% paraformaldehyde (Sigma-Aldrich) for 10 min, permeabilized in methanol, and blocked with 0.8% bovine serum albumin (Sigma-Aldrich) for 10 min. After incubation all night with primary antibodies against E-cadherin and N-cadherin, secondary antibodies were added for 2 h. Coverslips were mounted on glass slides after staining with DAPI (Sigma-Aldrich) for 10 min, and were finally examined using the TE2000-U microscope (Nikon, Tokyo, Japan). The experiment was conducted in triplicate.

Mitochondrial staining and mitochondrial fission analysis
SK-HEP-1 or Huh7 cells were transplanted onto coverslips and transfected. Cells were centrifuged for 5 min at room temperature for treatment with 0.1 μM of Mito-Tracker Red CMXRos (Molecular Probes, Thermo Fisher) for 30 min. After washing, images were taken using laser scanning confocal microscope. The experiment was conducted in triplicate.

Co-immunoprecipitation
Transfected cells were reaped from immunoprecipitation lysis buffer, centrifuged, and then cell lysates were incubated overnight with indicated antibodies at a constant speed at 4°C. Normal immunoglobulin G (IgG) was seen as a negative control. Following culturing with protein A-sepharose beads, the antigen-antibody mixture was washed three times with IP lysis buffer, eluted, and subjected to immunoblotting (Western blot). The experiment was conducted in triplicate.

Ubiquitination assay
Transfected cells were incubated in hot lysis buffer containing 1% SDS and 10 mM of N-ethylmaleimide (Sigma-Aldrich) and boiled at 100°C for 10 min. Cell lysates were diluted in SDS-free cell lysis buffer and mixed with anti-Flag-M2 agarose and assessed by western blotting. The experiment was conducted in triplicate.

Luciferase reporter assay
The wild-type or mutant sequences of c-Myb in LZTR1 promoter were sub-cloned into pGL3-basic vector (Promega, Madison, WI, USA) and co-transfected into 293T cells with pcDNA3.1/c-Myb or pcDNA3.1. The pGL3-LZTR1 promoter vector was co-transfected into SK-HEP-1 or Huh7 cells with sh-LL22NC03-N14H11.1#1/2 or sh-NC. Luciferase activities were studied with Dual-Luciferase reporter assay system (Promega), using Renilla luciferase as the internal control. The experiment was conducted in triplicate.

RNA pull-down assay
RNA pull-down assay was studied by Pierce Magnetic RNA-Protein Pull-Down Kit (Thermo Fisher Scientific, Waltham, MA, USA). Cell protein lysates of SK-HEP-1 or Huh7 were acquired by using RIPA lysis buffer, and then incubated with biotinylated RNA probes (LL22NC03-N14H11.1 biotin or LL22NC03-N14H11.1 no biotin). The streptavidin-coated magnetic beads (Invitrogen) were added to capture the RNA-protein mixture. Western blot was applied for analysis of the enrichment of proteins pulled down in indicated groups. The experiment was conducted in triplicate.

RNA immunoprecipitation
Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit was bought from Millipore (California, USA). Cell lysates were prepared using RNA immunoprecipitation (RIP) lysis buffer, and then cultivated with RIP buffer and magnetic beads conjugated to anti-c-Myb antibody, with anti-IgG antibody as a negative control. Besides, the interaction between U1 and SNRNP70 was used as a positive control for such assays. After adding proteinase K, precipitates were assayed by RT-qPCR. The experiment was conducted in triplicate.

Chromatin immunoprecipitation
Chromatin immunoprecipitation (ChIP) assay was undertaken with the application of EZ ChIP™ Chromatin Immunoprecipitation Kit (Millipore), following the standard method. Cells were treated in formaldehyde for 10 min for generating the DNA-protein cross-links. Then, cell lysates were sonicated to acquire chromatin fragments of 200-300 bp and immunoprecipitated with antibodies against c-Myb and control IgG. RT-qPCR was followed for the retrieved precipitated chromatin DNA. The experiment was conducted in triplicate.

Tumor xenograft model
Four-week-old male BALB/c nude mice were purchased from Shi Laike Company (Shanghi, China) and maintained in SPF-grade, pathogen-free animal lab. Tumor xenograft assay was performed via injecting nude mice subcutaneously with 1 × 10 6 SK-HEP-1 cells transfected with sh-LL22NC03-N14H11.1#1, sh-LL22NC03-N14H11.1#1 + sh-LZTR1, or sh-NC (mice were randomly divided into three groups). Tumor volume was recorded every 4 days. Four weeks later, mice were killed by cervical dislocation and tumors were weighted. Animal studies were approved by the Animal Ethics Committee of Affiliated Hospital of YouJiang Medical University For Nationalities.

TUNEL assay
To perform terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay in xenograft tissues, One-Step TUNEL Apoptosis Assay Kit was commercially acquired and used as instructed (Beyotime, Shanghai, China). After treatment with proteinase K for 15 min at room temperature, sections were fixed in 4% paraformaldehyde for 1 h and permeabilized with 0.1% Triton-X100 for 2 min, followed by incubation with TUNEL Assay Kit. The fluorescence microscope (Leica, Heerbrugg, Canton of St. Gallen, Switzerland) was utilized for observing the TUNEL-stained and DAPI-stained sections. The apoptotic nuclei were determined as the TUNEL and DAPI-positive nuclei located within tumor tissues. The experiment was conducted in triplicate.

Immunohistochemistry
The tumor tissues were acquired from xenograft assay and then fixed in 4% paraformaldehyde, dehydrated, and embedded in paraffin. Four-millimeter-thick sections were cut from paraffin-embedded xenograft tissues and deparaffinized, followed by cultivation with antibodies against Ki67 (Abcam), proliferating cell nuclear antigen (PCNA) (Abcam), E-cadherin and N-cadherin overnight at 4°C, and with biotinylated second antibody for 30 min at 37°C. The experiment was conducted in triplicate.

Hematoxylin and eosin staining
Tissues from xenograft model were immobilized for 24 h utilizing 4% formaldehyde and paraffin-embedded, followed by sectioned and stained with hematoxylin and eosin (HE) (Sigma-Aldrich). The optical microscope (Nikon) was applied for analyzing. The experiment was conducted in triplicate.

Statistical analysis
Data (in line with normal distribution) from assays conducted in triplicate were listed as mean ± SD. Statistical analysis was conducted by SPSS V. 19.0 (SPSS, Chicago, IL, USA) or Prism 6 (GraphPad Software, San Diego, CA, USA). The overall survival was plotted via the Kaplan-Meier method and compared by log-rank test. Gene expression correlation was analyzed with Pearson's method. Student's t test or analysis of variance was appropriately adopted for significant differences in groups, with P < 0.05 as threshold.

LL22NC03-N14H11.1 promoted HCC progression through inducing p-DRP1 (S616) and facilitating mitochondrial fission
A recent study reported that facilitated mitochondrial fission improved the survival and impeded apoptosis in HCC cells 19 . Moreover, studies have shown that inactivation of mitochondrial fission constrained cell migration and invasion in cancer 17,18 . Therefore, we wondered whether LL22NC03-N14H11.1 could regulate mitochondrial fission in HCC cells. MitoTracker Red staining analysis demonstrated that the mitochondrial elements were elongated and interconnected under LL22NC03-N14H11.1 knockdown or the treatment of Midiv-1, the selective inhibitor of DRP1, which is known to be the key regulator of mitochondrial fission (Fig. 3a). However, opposite phenomenon was observed in Hep3B and SNU-499 cells under enhanced expression of LL22NC03-N14H11.1 (Supplementary Fig. S2A). In addition, western blot analysis presented that LL22NC03-N14H11.1 knockdown reduced, whereas its upregulation boosted S616 phosphorylation of DRP1, while the total protein level of DRP1 was not changed under the above conditions (Fig. 3b, Supplementary Fig. S2B). Previous studies proved that p-DRP1 (S616) was required for the facilitation of mitochondrial fission and also for tumor growth 23,26 . Hence, these findings suggested that LL22NC03-N14H11.1 positively regulated mitochondrial fission in HCC cells through inducing p-DRP1 (S616). Later, we tried to examine whether LL22NC03-N14H11.1 influenced HCC progression through mitochondrial fission. As anticipated, it was proved that the hindered mitochondrial fission in LL22NC03-N14H11.1-silenced cells was normalized due to recovered p-DRP1 level upon DRP1 overexpression (Supplementary Fig. S2C, D). Thereafter, we confirmed that overexpression of DRP1 rescued the proliferation of HCC cells inhibited by sh- LL22NC03-N14H11.1#1 (Fig. 3c, d). Inductive effect of sh-LL22NC03-N14H11.1#1 on HCC cell apoptosis was abrogated by DRP1 overexpression (Fig. 3e). Invasion and migration of HCC cells restrained by sh-LL22NC03-N14H11.1#1 were restored by co-transfection of pcDNA3.1/DRP1 (Fig. 3f-g). The increase of E-cadherin and decrease of N-cadherin, MMP7, and MMP2 under LL22NC03-N14H11.1 knockdown were reversed by the overexpression of DRP1 (Fig. 3h). Jointly, it was indicated that LL22NC03-N14H11.1 promoted HCC progression through inducing p-DRP1 (S616)-facilitated mitochondrial fission.
Former studies stated that LZTR1 was a regulator of ubiquitination of RAS family, and both mutant and wildtype of RAS could be ubiquitinated and degraded by LZTR1 35 . We tried to examine whether LL22NC03-N14H11.1 influenced the regulation of LZTR1 on H-RAS (G12V). CoIP assays showed that in SK-HEP-1 cells, H-RAS (G12V) was enriched in the precipitates of anti-LZTR1, and knockdown of LL22NC03-N14H11.1 enhanced such enrichment, with the input H-RAS (G12V) level decreased and LZTR1 level increased (Fig. 4d). Also, the ubiquitination of H-RAS (G12V) was induced under the knockdown of LL22NC03-N14H11.1, with the input H-RAS (G12V) level decreased and LZTR1 level increased (Fig. 4e). These results indicated that LL22NC03-N14H11.1 reduced the interaction between LZTR1 and H-RAS (G12V) by reducing LZTR1 expression. We confirmed through western blot that knockdown of LL22NC03-N14H11.1 decreased H-RAS (G12V) level and increased LZTR1 level, and such results could be reversed by the knockdown of LZTR1 in HCC cells (Fig. 4f). Additionally, silence of LL22NC03-N14H11.1A induced the mRNA level of LZTR1, but knockdown of LZTR1 had no impact on the level of LL22NC03-N14H11.1 in HCC cells (Fig. 4g, h). We then found that LZTR1 expression was downregulated and negatively correlated with LL22NC03-N14H11.1 expression in HCC tissues (Fig. 4i, j). The low expression of LZTR1 in HCC cell lines vs. normal cell line was also verified (Fig. 4k). Collectively, LL22NC03-N14H11.1 inhibited LZTR1mediated ubiquitination of H-RAS (G12V) and activated MAPK pathway to induce p-DRP1 (S616).

Discussion
Studies have provided convincing evidences that hepatocarcinogenesis involves dysregulation of lncRNAs. Large amounts of lncRNAs are identified to possess prognostic and diagnostic values in HCC 8,9 . Therefore, identifying novel lncRNAs in HCC and finding out their modulatory mechanism can benefit the progress of treatment efficacy in HCC and improve the survival of HCC patients.
This study analyzed the dysregulated lncRNAs in HCC through two bioinformatics tools (circlncRNAnet and GEPIA) and found that LL22NC03-N14H11.1 was a new lncRNA highly expressed in HCC samples and had great prognostic significance for HCC patients. We also confirmed that LL22NC03-N14H11.1 level was elevated in HCC tissues and cell lines. Besides, we found that LINC00152 was also highly expressed in HCC samples. A number of former studies have elucidated that LINC00152 served as an oncogene in HCC [36][37][38] , but the role of LL22NC03-N14H11.1 in HCC has never been investigated. Therefore, we focused on the exploration of LL22NC03-N14H11.1 in HCC. Functionally, we discovered that LL22NC03-N14H11.1 accelerated proliferation, migration, invasion, and EMT in HCC cells. These findings indicated that targeting LL22NC03-N14H11.1 in HCC might be a promising approach to relieve hepatocarcinogenesis.
Mitochondrial fission has been demonstrated as an important activity related to cell survival and metastasis in tumors. Some works stated that increased mitochondrial fission contributes to drug-induced apoptosis and enhance chemo-sensitivity in cancer cells 45 , whereas others argued that inhibiting mitochondrial fission can retard cell proliferation and metastasis in several cancer types, such as colon cancer, lung cancer, and breast cancer [14][15][16][17][18] . A recent study observed that mitochondrial fission was facilitated in HCC samples compared to adjacent normal samples, and that DRP1 inhibition could inhibit cell survival and autophagy in HCC 19 , indicating that excessive mitochondrial fission in HCC was oncogenic. In this study, we found that LL22NC03-N14H11.1 silence prevented mitochondrial fission, same as the effect of Midivi-1 (DRP1 inhibitor). Also, we validated that overexpressing DRP1 could rescue the inhibitory effect of LL22NC03-N14H11.1 silence on HCC progression, indicating that LL22NC03-N14H11.1 regulated HCC through facilitating DRP1-regulated mitochondrial fission.
DRP1 is a well-known member of dynamin-related GTPases, which were referred to as primary regulators of mitochondrial dynamics 13,20 . It has been well established that phosphorylation of DRP1 at S616 led to DRP1 activation and facilitated mitochondrial fission 21 . Herein, we found that LL22NC03-N14H11.1 contributed to DRP1 phosphorylation at S616, with no impact on the level of total DRP1. Previous works have identified several regulators of p-DRP1 (S616), such as CDK1, DPI1, and ERK1/2 [22][23][24] . Herein, we discovered that among the aforementioned regulators, only ERK1/2 phosphorylation was positively regulated by LL22NC03-N14H11.1 in HCC cells. ERK1/2 was the key regulators in ERK/MAPK pathway. It is widely known that MAPK pathway is a key signaling related to the survival, apoptosis, and metastasis in tumor cells 46,47 . Recent studies revealed that H-RAS (G12V) activated MAPK pathway so that ERK1/2 was phosphorylated and induced p-DRP1 (S616) 25 . Importantly, such mechanism has been proved to accelerate mitochondrial fission and tumor growth in pancreatic cancer 26 . Hence, it was reasonable to deduce that LL22NC03-N14H11.1 regulated MAPK pathway to induce mitochondrial fission in HCC.
Expectedly, we firstly found that LL22NC03-N14H11.1 knockdown reduced H-RAS (G12V) level in HCC cells. We further suggested that LL22NC03-N14H11.1 regulated H-RAS (G12V) protein stability. Several studies have Fig. 7 LL22NC03-N14H11.1 drove tumorigenesis and metastasis in HCC through LZTR1 in vivo. SK-HEP-1 cells were, respectively, transfected with sh-NC, sh-LL22NC03-N14H11.1#1, or sh-LL22NC03-N14H11.1#1 + sh-LZTR1 and injected subcutaneously or from tail vain to monitor tumorigenesis and metastasis of HCC. a Volumes of xenografts in mice of each group was evaluated every 4 days after subcutaneous injection and the growth curve was outlined. b Twenty-eight days after injection, tumors from each group were resected and weighed. c TUNEL staining was used to evaluate the apoptosis in tumors of each group. d RT-qPCR analysis of LL22NC03-N14H11.1 and LZTR1 in tumors of each group. e Western blot was implemented to detect the levels of LZTR1, c-Myb, H-RAS (G12V), p-ERK1/2, total ERK1/2, p-DRP1 (S616), and total DRP1 in tumors of each group. f, g IHC staining and quantification of Ki67, PCNA, N-cadherin, and E-cadherin in tumors of each group. h Metastatic nodules in tumors of each group was stained by HE and quantified. Scale bar: 100 μm. **P < 0.01; n.s. no significance.
pointed out that LZTR1 was a conserved regulator of the ubiquitination of RAS family. LZTR1 was suggested to possess tumor-suppressing function in cancers and be related to cell apoptosis [32][33][34] . A study elucidated that LZTR1 ubiquitinated K-RAS, M-RAS, N-RAS, H-RAS, and their mutant types, leading to the inactivation of MAPK pathway 35 . Therefore, it was reasonable to suggest that LZTR1 exerted tumor-suppressive effect through suppressing MAPK pathway, and that LZTR1 might mediate the regulation of LL22NC03-N14H11.1 on H-RAS (G12V) in HCC. This study firstly revealed that LL22NC03-N14H11.1 reduced LZTR1 expression to inhibit the LZTR1-mediated ubiquitination of H-RAS (G12V).
Furthermore, we uncovered that LL22NC03-N14H11.1 repressed LZTR1 expression at the transcriptional level. We identified that LL22NC03-N14H11.1 was mainly located in the nucleus of HCC cells. The mechanisms whereby lncRNAs regulate target genes are different depending on their cellular localization 10,11 . In the nucleus, lncRNAs can recruit certain transcription factors to affect the transcription of target genes 40,41 . For instance, lncRNA REG1CP could tether FANCJ to REG3A promoter to induce REG3A activation by interacting with FANCJ in colorectal cancer 48 . Also, a latest research discovered that lncRNA Oplr16 could activate Oct4 via inducing DNA demethylation by recruiting TET2 49 . Herein, we identified that c-Myb potentially targeted LZTR1 promoter and interacted with LL22NC03-N14H11.1. As a transcription factor, c-Myb is demonstrated to either suppress or activate gene transcription 42,43 , and it has been proved to contribute to hepatocarcinogenesis and metastasis in HCC 50 . In this study, we validated that LL22NC03-N14H11.1 recruited c-Myb to LZTR1 promoter so as to repress LZTR1 transcription and activate H-RAS/MAPK pathway, suggesting LL22NC03-N14H11.1 as a scaffold in HCC as many other lncRNAs do 44,51 . Besides, it has been revealed that c-Myb could repress transcription by recruiting certain transcription repressor or competing with other transcription activators for target promoters 43 . However, the precise mechanism whereby c-Myb inhibited LZTR1 transcription remains yet to be further investigated in the future. Finally, we validated that LL22NC03-N14H11.1 promoted HCC tumorigenesis and metastasis in vivo.