Abstract

   Hepatocellular carcinoma (HCC) is a deadly malignancy with limited
   treatment options. As a first-line treatment for advanced HCC,
   Lenvatinib has been applicated in clinic since 2018. Resistance to
   Lenvatinib, however, has severely restricted the clinical benefits of
   this drug. Therefore, it is urgent to explore the potential resistance
   mechanisms of Lenvatinib and identify appropriate methods to reduce
   resistance for the treatment of HCC. We identified SAHA, a HDAC
   inhibitor, to have effective anti-tumor activity against
   Lenvatinib-resistant HCC organoids by screening a customized drug
   library. Mechanism analysis revealed that SAHA upregulates PTEN
   expression and suppresses AKT signaling, which contributes to reversing
   Lenvatinib resistance in liver cancer cells. Furthermore, combinational
   application of Lenvatinib and HDAC inhibitor or AKT inhibitor
   synergistically inhibits HCC cell proliferation and induces cell
   apoptosis. Finally, we confirmed the synergistic effects of Lenvatinib
   and SAHA, or AZD5363 in primary liver cancer patient derived organoids.
   Collectively, these findings may enable the development of Lenvatinib
   combination therapies for HCC.

   Keywords: Hepatocellular carcinoma, Lenvatinib resistance, SAHA, AKT,
   combination treatment

Introduction

   Hepatocellular carcinoma (HCC), as the most frequent primary liver
   cancer, is ranked as the sixth most common neoplasm and the third
   leading cause of cancer-related death worldwide [37]^1. Standard
   treatments for early-stage HCC comprise surgical resection,
   transplantation, and ablation [38]^2. However, for advanced HCC
   patients who are ineligible for surgical resection, systemic therapies
   such as Sorafenib [39]^3, Lenvatinib [40]^4, or Atezolizumab plus
   Bevacizumab [41]^5 are commonly applicated. Lenvatinib, a multi-kinase
   inhibitor, was approved as a first-line treatment for patients with
   advanced HCC [42]^4. Lenvatinib targets multiple kinase receptors,
   including VEGFR1-3, FGFR1-4, PDGFRα, RET, and KIT [43]^6^, [44]^7.
   Studies have demonstrated that Lenvatinib offers comparable overall
   survival to Sorafenib in untreated advanced HCC, but with better
   progression-free survival [45]^4. However, despite a small number of
   patients obtained a real and long-term benefit from this therapy, most
   patients do not respond or gradually develop resistance over time to
   Lenvatinib treatment [46]^8. Therefore, it is crucial to acquire a
   comprehensive understanding of the mechanisms of Lenvatinib resistance
   and explore new therapeutics to overcome it.

   Epigenetic regulation plays a crucial role in controlling tumor
   initiation and progression [47]^9^, [48]^10. Histone deacetylases
   (HDACs) are essential for chromatin remodeling and regulation of gene
   transcription, while altered expression of HDACs actively contributes
   to tumor initiation and progression [49]^11. Furthermore, HDAC
   inhibitors represent the pioneering success of epigenetics-based cancer
   therapy [50]^12^-[51]^15. The FDA has approved multiple HDACs-targeting
   drugs for cancer treatment. So far, four HDAC inhibitors, including
   SAHA, Romidepsin, Belinostat, and Panobinostat, have received FDA
   approval [52]^16. For example, SAHA and Panobinostat have been approved
   to treat cutaneous T-cell lymphoma and multiple myeloma, respectively
   [53]^17^-[54]^20. Meanwhile, numerous HDAC inhibitors have been
   characterized to effectively suppress tumor growth and metastasis in
   both in vitro and in vivo models [55]^21. Collectively, recent clinical
   and basic studies have indicated promising potentials of HDAC
   inhibitors as novel anticancer drugs [56]^22.

   The PI3K/AKT signaling pathway governs essential cellular processes,
   such as cell cycle regulation, survival, metabolism, motility, and
   angiogenesis, in physiological and pathological settings and is
   implicated as one of the key cancer hallmarks [57]^23^, [58]^24.
   Dysregulation of the PI3K/AKT signaling pathway is frequently observed
   in HCC, warranting further investigation and prompting the development
   of novel therapeutic targets for liver cancer [59]^25^, [60]^26. The
   PI3K/AKT pathway plays a critical role in mediating chemoresistance in
   multiple cancer types, including breast cancer, leukemia, lung cancer,
   ovarian cancer, melanoma, and hepatocellular carcinoma [61]^27^,
   [62]^28. Additionally, activation of the PI3K/AKT pathway has been
   reported to promote acquired resistance to Sorafenib in HCC cells
   [63]^29. Utilizing the currently available small molecule inhibitors
   targeting AKT presents a potential strategy to overcome tumor therapy
   resistance [64]^28.

   In this study, we found HDAC inhibitor SAHA could effectively suppress
   Lenvatinib-resistant HCC organoids' survival. Both in vivo and in vitro
   studies demonstrated that SAHA inhibited HCC cell proliferation,
   induced cell apoptosis, and enhanced the sensitivity of
   Lenvatinib-resistant HCC to Lenvatinib treatment via inhibiting the AKT
   signaling pathway. In conclusion, our findings suggest that a
   combination therapy of SAHA and Lenvatinib could enhance the
   therapeutic efficacy and impede the progression of HCC.

Materials and Methods

Cell lines

   The human HCC cell lines, HCCLM3, SK-Hep1, HepG2, SNU-739 and the mouse
   HCC cell line Hepa1-6 were obtained from Key Laboratory of Cancer
   Prevention and Therapy, Tianjin Medical University Cancer Institute and
   Hospital, Tianjin Medical University (Tianjin, China). The Human HCC
   cell lines Hep3B, Huh7 and PLC/PRF/5 cell lines were kindly provided by
   Dr. Zhiqian Zhang from Peking University Cancer Hospital and Institute
   (Beijing, China). All Cell lines were validated by STR analysis
   (BIOWING, Shanghai, China). All human HCC cells and mouse liver cancer
   cells were cultured in Dulbecco's modified Eagle medium supplemented
   with 10% fatal bovine serum and 1% penicillin-streptomycin at 37 °C in
   a humidified atmosphere with 5% CO[2]. Mycoplasma contamination was
   excluded via a polymerase chain reaction-based method.

Cell viability assays

   HCC cells were plated in 96 well plates at optimized cell density. The
   day after cell seeding, cells were treated with compounds at indicated
   concentrations for 72 hours. All compounds used for cell viability
   assays were purchased from MedChemExpress (Shanghai, China). Cell
   viability was measured by sulforhodamine B assay as described
   previously [65]^30. All experiments were performed in triplicate for at
   least twice.

Long-term clonogenic assays

   Cells were plated into 6-well plates at a density of 1-3×10^3 cells per
   well. The day after cell seeding, cells were cultured in medium
   containing indicated drugs for 7-14 days with changing culture medium
   twice a week. Cells were fixed with 4% formaldehyde and stained with
   0.1% crystal violet diluted in water. Crystal Violet was solubilized
   with 10% acetic acid, and optical results were read by an automated
   spectrophotometric plate reader at a wavelength of 595 nm.

Cell apoptosis analysis

   Drug treated and control cells were collected and washed by 1×PBS once.
   1×10^5 cells were stained with anti-Annexin V antibody (eBiosciences,
   Vienna, Austria) and propidium iodide following the manufacturer's
   protocol. Briefly, cells were washed with 1×binding buffer followed by
   staining with FITC-labeled annexin V in dark for 30 min at room
   temperature. Subsequently, the cells were washed with 1×binding buffer
   twice, and stained with propidium iodide. Stained cells were analyzed
   on the BD FACSCalibur (BD bioscience, NJ, USA).

RNA sequencing and data analysis

   SAHA or vehicle control treated SNU-739 cells were collected using
   TRIzol (Invitrogen, CA, USA) for RNA extraction and sequencing. Library
   preparation and RNA-seq were performed by Novogene (Beijing, China)
   using paired-end sequencing pipeline. Paired-end reads were generated
   on Illumina Hiseq 2500 platform. After quality controlling, clean data
   were aligned to UCSC hg19 reference by STAR and quantified using RSEM
   with default paraments. Differentially expressed genes were determined
   with the R package DESeq2. KEGG analysis were performed using
   ClusterProfiler package.

AKT overexpression and siRNA transfection

   The lentiviral pCDH-AKT plasmid was kindly provided by Dr. Ceshi Chen
   from Kunming Medical University (Kunming, China), and virus packaging
   was performed according to their published protocols [66]^31.
   Lentiviruses infected Hep3B and Huh7 cells were selected by 1 μg/ml
   puromycin for further experiments. siRNAs against AKT were designed and
   synthesized by JTS Scientific (Beijing, China). The siRNA sequences are
   shown in [67]Table S1. SNU-739 and Hepa1-6 cells were transfected with
   20 nmol/L siRNA following the general transfection protocol. 36 hours
   after transfection, the cells were harvested for further experiments.

Drug synergy analysis

   Cells were plated in 96-well plates at 1×10^3 cells/well, and 24 hours
   later, were treated with single or combined compounds as designed for
   72 hours. Cell viability was measured by SRB assays. We evaluated 6
   doses of Lenvatinib combined with 6 doses of SAHA (36 dose
   combinations) in triplicate, synergy score was analyzed and visualized
   with the online software Synergy Finder (version 2.0) using the Bliss
   model.

Quantitative real-time PCR

   Total RNA from HCC cells was isolated following TRIzol RNA extraction
   protocol and reverse-transcribed to cDNA using a RT-PCR kit (TianGen,
   Beijing, China). Quantitative PCR analysis was performed on the AriaMx
   Real-Time PCR System (Agilent Technologies, CA, USA) using SYBR Green
   PCR reagent (TianGen, Beijing, China) following the manufacturer's
   instructions. The sequences of primers used in our study are listed in
   [68]Table S1.

Western blotting

   Whole cell lysates were prepared from cultured cells or organoids, and
   30μg proteins per sample were loaded for gel separation and transferred
   to the PVDF membrane. The membranes were incubated with primary
   antibodies at 4 °C overnight. Then, the blots were incubated with the
   horseradish peroxidase-conjugated secondary antibody and developed by
   enhanced chemiluminescence. β-actin was detected as the internal
   control. Antibodies against PTEN, p-AKT, AKT, Cyclin D1, P21, p-BAD,
   BAD, Acetyl-H3, H3, and PARP were purchased from Cell Signaling
   Technology (MA, USA). Antibodies against β-actin, p-ERK1/2 and ERK1/2
   were purchased from Santa Cruz biotechnology (CA, USA).

ChIP assay

   After SAHA or DMSO treatment, HCC cells were cross-linked with 1%
   formaldehyde for 10 minutes at room temperature, then lysed in SDS
   lysis buffer with protease inhibitors. The cell lysates were sonicated
   with Q125 Sonicator (QSonica, CT, USA) to obtain chromatin fragments of
   300-1000 bp. Specific antibodies recognizing acetyl-histone H3K9 were
   employed to interact and pulldown corresponding protein-DNA fragments
   using protein A/G beads. The immunoprecipitated DNA was purified for
   further quantitative PCR analysis. The primers targeting the PTEN
   promoter region for qPCR assay are listed in [69]Table S1.

ATAC-Seq

   The ATAC assay was conducted on SNU-739 cells treated with either DMSO
   or SAHA (5 μM) for 24 hours according to the manufacturer's protocol,
   utilizing the Hyperactive ATAC-Seq Library Prep Kit for Illumina
   (Vazyme, Nanjing, China). Subsequently, the ATAC‐seq libraries
   underwent paired-end sequencing on the HiSeq-PE150 platform. The
   ATAC-seq reads were aligned to the hg38 reference genome using Bowties,
   and the visualization of the PTEN locus was performed with IGV 2.17.2
   software.

In vivo mouse syngeneic graft experiments

   All mice were purchased from SiPeiFu Biotechnology (Beijing, China) and
   housed in a 12-hour light/dark cycle with access to food and water ad
   libitum. For the syngeneic graft model, 5×10^5 Hepa1-6 cells were
   implanted to 6-week-old C57BL/6 mice subcutaneously. Once tumors
   reached approximately 50 mm^3, mice were randomized into designed
   groups (6 mice per group) for drug treatment as indicated. Tumor
   volumes were recorded every 3 days and calculated with the formula
   Volume (mm^3) = L×W^2×0.5 (L is the longest tumor axis and W is the
   shortest tumor axis). At the end of treatment, the mice were sacrificed
   and tumors were harvested for further analysis.

Immunohistochemistry staining

   The murine tumor samples were collected for Ki67 and cleaved caspase-3
   detection using a standard diaminobenzidine (DAB) immunohistochemistry
   staining protocol as described previously. In brief, antigen-retrieved
   tissue slides were incubated with designed primary antibodies at 4 °C
   overnight followed by incubating with the biotinylated anti-mouse
   secondary antibody and the avidin-biotin complex (Gene Tech, Shanghai,
   China) sequentially. Finally, slides were incubated with DAB (Gene
   Tech, Shanghai, China) until suitable staining developed.
   Immunostaining images were taken under Axioscope 5 microscope (Carl
   Zeiss, Jena, Germany). Antibody against cleaved caspase-3 was purchased
   from Cell Signaling Technology (MA, USA). The antibody against Ki-67
   was purchased from Abcam (MA, USA).

Organoid culture and drug response assay

   HCC samples used in this study were obtained during liver resection
   performed in Tianjin Cancer Hospital (Tianjin, China), and Henan Cancer
   Hospital (Zhengzhou, China). All patient samples in this study were
   collected with informed consent in accordance with the Declaration of
   Helsinki. The study was approved by hospital ethics committees
   (BC2020096 and 2016CT054). Detailed clinicopathological data are
   summarized in [70]Table S2. Organoids were cultured according to
   protocols as described previously [71]^32. Briefly, fresh liver cancer
   tissues were cut into small pieces, followed by incubating in DMEM/F12
   containing 2mg/ml collagenase D and 0.1 mg/ml DNase I for 1 hour at 37
   °C, with mechanical pipetting every 15 minutes. Digested cells (most in
   clusters) were filtered through 100 μm cell strainers and collected for
   culture. Cells were suspended in 60% matrigel diluted by complete
   culture media and plated in 50 μl drops into 6-well non-treated tissue
   culture plates and incubated for 30 min at 37 °C for gel
   solidification. Subsequently, 2 mL organoid culture media was added,
   and the medium was changed every 3 days.

   For drug screening assays, collected organoids were digested by 1mg/ml
   dispase Ⅱ to remove Matrigel, followed by passing through a 100 μm cell
   strainer to remove extra-large organoids. Subsequently, organoids were
   resuspended in complete organoid culture medium (2×10^4 organoids/ml)
   supplemented with 5% matrgel and plated into ultralow-attachment
   96-well plates in triplicate. 24 h after plating, drugs were added as
   designed at indicated concentrations. 6 days after drug treatment, cell
   viability was analyzed using CellTiter-Glo 3D reagent (Promega, WI,
   USA) according to the manufacturer's instructions.

Statistical analysis

   Statistical analysis was performed using SPSS, version 26.0. The data
   are shown as mean ± SD. The significance was analyzed by Student's
   t-test. P values < 0.05 were considered statistically significant.

Results

Drug screening identifies SAHA as a potential candidate for overcoming
Lenvatinib-resistance in HCC

   In order to find drugs that can overcome Lenvatinib resistance, we
   selected 48 compounds, which have at least passed Phase I clinical
   trials with anti-cancer potentials from clinicaltrials.org, and
   performed drug screening using two Lenvatinib-resistant liver cancer
   organoids. The results showed that the HDAC inhibitor, SAHA, exhibited
   surprisingly effective inhibitory effects on these primary liver cancer
   organoids (Fig. [72]1A and S1A). We subsequently performed
   dose-response tests on 8 liver cancer cell lines with Lenvatinib and
   SAHA. As the results shown in figures [73]1 and S1, these HCC cell
   lines showed IC[50]s ranging from 0.601 μM to 12.55 μM to Lenvatinib
   treatment, while showed an average IC[50] of 1.251±0.426 μM to SAHA,
   indicating these HCC cell lines are all relatively sensitive to SAHA
   regardless their response to Lenvatinib. We then confirmed the effects
   of SAHA on HCC cells using long-term clonogenic assays (Fig. [74]1C,
   1D, S1E and S1F) and got similar results that SAHA could suppress HCC
   clone formation effectively at relatively low doses.

Figure 1.

   [75]Figure 1
   [76]Open in a new tab

   Drug screening identifies SAHA as an effective inhibitor of HCC. (A)
   Drug screening on two Lenvatinib-resistant HCC organoids. Organoids
   viability was assessed on Day 6 posttreatment using each compound at a
   concentration of 10 μM. (B) Dose-response curve of SNU-739 and Hepa1-6
   cells treated with SAHA. (C) Colony formation of SNU-739 and Hepa1-6
   cells in present or absent of SAHA treatment. (D) The quantification of
   the three independent assays shown in (C). Data are presented as mean ±
   SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, by two-sided Student's
   t-test). (E) Apoptosis induced by SAHA in SNU-739 and Hepa1-6 cells.
   Cell lines were treated as indicated for 24 hours, after which annexin
   V/PI staining was performed, followed by flow cytometry. (F) The
   quantification of the three independent assays shown in (E). Data are
   presented as mean ± SD (n = 3, *P < 0.05, **P < 0.01, ***P < 0.001, by
   two-sided Student's t-test). (G) Western blotting plots of PARP and
   cleaved PARP in SNU-739 and Hepa1-6 cells treated with SAHA.

   Based on the cell viability assays, we selected 2 HCC cell lines,
   SNU-739 and Hepa1-6, with relatively high IC[50]s to Lenvatinib
   treatment for further analysis ([77]Fig. S1B and S1C). Considering both
   cell growth inhibition and cell death, including cell apoptosis,
   contribute to cell viability reduction, we also checked the effects of
   SAHA on HCC cell apoptosis. As shown in Figure [78]1E-[79]1G, SAHA
   efficiently induced cell apoptosis in both SNU-739 and Hepa1-6 cells as
   indicated by increased Annexin V^+ cells (Fig. [80]1E and [81]1F) and
   increased expression of cleaved PARP (Fig. [82]1G). In summary, these
   findings indicate that SAHA has a good inhibitory effect on
   Lenvatinib-resistant liver cancer cells.

SAHA sensitizes HCC cells to Lenvatinib treatment

   Since Lenvatinib resistant HCC cells are relatively sensitive to HDACi
   treatment, we wondered whether HDAC inhibition could help address the
   Lenvatinib resistance issues in these HCC cells. Lenvatinib, together
   with SAHA, were employed to treat these Lenvatinib resistant HCC cells
   (SNU-739 and Hepa1-6), and the synergy scores were calculated using
   Synergyfinder 2.0. Interestingly, application of low dose SAHA (0.5 μM)
   significantly sensitized these cells to 2.5 μM Lenvatinib treatment
   (Fig. [83]2A), the synergy scores of SAHA and Lenvatinib in SNU-739 and
   Hepa1-6 cells reached 17.83 and 27.52, respectively (Fig. [84]2B). In
   addition, except for the SK-Hep1 cells, the synergy scores of SAHA and
   Lenvatinib in HCCLM3, HepG2, and PLC/PRF/5 cells were all greater than
   10, indicating a significant synergistic effect ([85]Fig. S2A).

Figure 2.

   [86]Figure 2
   [87]Open in a new tab

   Lenvatinib and SAHA demonstrates synergistic effects in HCC. (A) Growth
   curve of SNU-739 and Hepa1-6 cells treated with either DMSO, Lenvatinib
   (2.5 μM), SAHA (0.5 μM) or their combination. (B) Synergy map of
   SNU-739 and Hepa1-6 cells treated with indicated concentrations of SAHA
   and Lenvatinib. (C) Colony formation of SNU-739 and Hepa1-6 cells
   treated with either DMSO, Lenvatinib (2.5 μM), SAHA (0.5 μM) or their
   combination. (D) The quantification of the three independent assays
   shown in (C). (E) Cell apoptosis analysis of SNU-739 and Hepa1-6 cells
   treated with either DMSO, Lenvatinib, SAHA or their combination. (F)
   The quantification of the three independent assays shown in (E). (G)
   Tumor growth curve of Hepa1-6 xenograft models. Mice were treated with
   vehicle, Lenvatinib (4 mg/kg), SAHA (40 mg/kg), or combination daily
   for 18 days. Growth curve was plotted by measuring the relative tumor
   volume every 3 days. (H-I) The excised tumors and relative tumor weight
   at the termination day from the experiment described in (G),
   respectively.

   To further confirm whether SAHA can overcome Lenvatinib resistance in
   liver cancer cells, we employed cell growth curve analysis and
   long-term cloning formation assay in multiple Lenvatinib resistant
   cells to evaluate the combinatorial therapeutic effects of SAHA and
   Lenvatinib. Indeed, the combination application of SAHA and Lenvatinib
   significantly inhibited cell proliferation and cloning formation
   ability (Fig. [88]2C-[89]2D and S2B-2D). Meanwhile, we identified SAHA
   and Lenvatinib synergistically promote apoptosis in both SNU-739 and
   Hepa1-6 cells (Fig. [90]2E-F and S2E).

   Furthermore, using in vivo xenograft transplanting mouse model, we
   found combined use of SAHA and Lenvatinib significantly inhibited
   Hepa1-6 cell growth subcutaneously in nude mice compared with each
   compound alone (Fig. [91]2G-[92]2I), without reducing the body weight
   of mice significantly ([93]Fig. S2F). Immunohistochemical analysis of
   these tumor mass showed that combined drug treatment significantly
   suppressed Ki67 expression, and increased cleaved caspase-3 expression
   ([94]Fig. S2G and S2H), indicating decreased proliferation and
   increased cell death of these tumors in vivo. Together, these results
   suggested that the combinatorial application of SAHA and Lenvatinib
   synergistically inhibits HCC tumor cell proliferation and promote cell
   apoptosis, both in vitro and in vivo.

SAHA suppresses PI3K-AKT signaling by up-regulating PTEN expression

   In order to better understand the anti-tumor mechanisms of SAHA in
   Lenvatinib resistant HCC, SNU-739 cells were treated with SAHA, and
   RNA-sequencing were performed to detect genes, pathways and regulatory
   networks that were dysregulated. Transcriptome analysis revealed that
   1506 genes were downregulated after SAHA treatment (Fig. [95]3A). KEGG
   pathway enrichment analysis identified the PI3K-AKT signaling pathway
   as the most downregulated pathway in SAHA-treated SNU-739 cells (Fig.
   [96]3B). Detailed analysis of PI3K-AKT signaling revealed genes,
   including PTEN and CDKN1A are upregulated, while CCND1 is downregulated
   in this pathway ([97]Fig. S3A). We validated these genes' expression
   using RT-PCR in SAHA-treated SNU-739 and Hepa1-6 cells, and confirmed
   these genes had the same changing trends in both cell lines (fig.
   [98]3C).

Figure 3.

   [99]Figure 3
   [100]Open in a new tab

   SAHA inhibits the AKT pathway by upregulating PTEN expression. (A)
   volcano plot showing differentially expressed genes in SNU-739 cells
   post SAHA treatment (5 μM) for 24 hours. (B) Pathway enrichment
   analysis of down-regulated genes in SAHA treatment cells compared to
   DMSO treatment cells. (C) Real-time PCR analysis of the expression of
   PTEN, CCND1 and CDKN1A in SAHA treatment cells. (D) Histone acetylation
   in PTEN promoter. SNU-739 cells were treated with SAHA (5 μM) for 12
   hours before being subjected to ChIP assay using anti-acetyl-histone
   H3K9 (Ac-H3K9) antibody followed by qPCR analysis using primers
   targeting indicated PTEN promoter region. (E-F) Western blotting
   analysis of the downstream of AKT signaling expression in SNU-739 and
   Hepa1-6 cells, which were treated with SAHA in a dose- and
   time-dependent manner. β-actin served as a loading control.

   It's well-documented that HDAC mainly works through removing acetyl
   groups from DNA-binding histone proteins, which is generally associated
   to a decrease in chromatin accessibility for transcription factors and
   repressive effects on gene expression. We then checked whether SAHA
   promotes target genes' expression, such as PTEN, by opening chromatin
   structure. We performed chromatin immunoprecipitation (ChIP) assay
   using acetyl-histone H3K9 antibody in SAHA-treated SNU-739 cells,
   followed by qPCR analysis. As expected, SAHA treatment significantly
   enriched the presence of the PTEN gene promoter region in the
   precipitant pulled down by Ac-H3K9, as probed by three independent
   pairs of primers specifically targeting PTEN gene promoter, indicating
   SAHA treatment increased histone acetylation in PTEN promoter region
   (Fig. [101]3D). To further confirm the impact of SAHA on PTEN
   transcription regulation, we utilized ATAC-Seq to analyze the chromatin
   accessibility of the PTEN promoter region in SNU-739 cells after SAHA
   treatment. Regions of accessible chromatin surrounding PTEN were
   increased by SAHA treatment, especially in the promoter region
   ([102]Fig. S3B). SAHA is a pan-inhibitor for multiple HDACs, especially
   subtype I (HDAC1, 2, 3, and 8), IIa (HDAC4, 5, 7, and 9), IIb (HDAC6,
   and 10) and IV (HDAC11) [103]^33. To investigate which HDAC is
   responsible for regulating PTEN expression, we depleted HDAC1-11 in
   SNU-739 cells using siRNAs targeting each HDAC respectively. As the
   results shown in [104]supplementary figures 3C-E, PTEN expression is
   significantly increased upon silencing either HDAC1, HDAC3, HDAC4, or
   HDAC7, while no obvious effects were observed when other HDACs were
   depleted, indicating SAHA may suppress PTEN expression by targeting one
   or multiple members of HDAC1, HDAC3, HDAC4, and HDAC7. Moreover,
   immunoblotting results also showed SAHA treatment led to attenuated ATK
   phosphorylation, Cyclin D1 expression, and increased P21 expression in
   a dose- and time-dependent manner (Fig. [105]3E and [106]3F). These
   data suggest SAHA treatment inhibits the AKT signaling pathway in HCC
   cells.

Activated PI3K-AKT pathway contributes to Lenvatinib resistance of HCC

   Aberrant activation of PI3K-AKT pathway is associated with
   tumorigenesis, cancer progression, and drug resistance in various types
   of malignancies, and is considered as one of the most effective targets
   for cancer therapy [107]^28. In order to investigate whether PI3K-AKT
   signaling contributes to Lenvatinib resistance of HCC, we first
   conducted transcriptome analysis using publicly available data
   ([108]GSE211850). Compared to Huh7 parental cells, we observed
   enrichment of the PI3K/AKT signaling pathway in Lenvatinib-resistant
   Huh7 cells ([109]Fig. S4A and S4B). Subsequently, we examined the
   expression of AKT in HCC cell lines, and found that AKT and p-AKT
   expression were relatively higher in Lenvatinib-resistant cell lines
   compared to sensitive ones (Fig. [110]4A, S1B and S1C). We further
   ectopically overexpressed the AKT in Lenvatinib-sensitive Hep3B and
   Huh7 cell lines (Fig. [111]4B and S4C), and found AKT overexpressed
   cells are more resistant to Lenvatinib treatment (Fig.
   [112]4C-[113]4F).

Figure 4.

   [114]Figure 4
   [115]Open in a new tab

   Activation of the AKT pathway leads to resistance to Lenvatinib in HCC.
   (A) Western blotting analysis of the expression of total AKT and p-AKT
   in a panel of liver cancer cell lines. (B) Western blotting analysis of
   AKT and p-AKT expression in Hep3B and Huh7 cells with transduction of
   pCDH-AKT lentiviral particles. (C) Dose-response curves of Hep3B and
   Huh7 cells with vector or AKT overexpression treated with Lenvatinib.
   (D) Growth curve of control or AKT overexpressed Hep3B and Huh7 cells
   treated with Lenvatinib (2.5 μM). (E) Colony formation assay in Hep3B
   and Huh7 cells treated with Lenvatinib after overexpressed AKT or
   vector. (F) The quantification of the three independent assays shown in
   (E).

   Since AKT activation contributes to Lenvatinib-resistance in HCC cells,
   we wondered whether blocking such signaling might facilitate overcoming
   HCC cell resistance to Lenvatinib. Interestingly, AKT depletion (Fig.
   [116]5A and S4D) significantly enhanced SNU-739 and Hepa1-6 cells'
   sensitivity to Lenvatinib treatment (Fig. [117]5B). Cell proliferation
   and long-term colony formation assays yielded consistent results (Fig.
   [118]5C-E). As aforementioned, AKT overexpression conferred resistance
   to Lenvatinib in Hep3B and Huh7 cells, however, treatment with SAHA
   markedly decreased AKT phosphorylation level ([119]Fig. S4E and S4F)
   and partially restored sensitivity to Lenvatinib ([120]Fig. S5A-5C),
   which further confirmed AKT inhibition sensitizes HCC cells to
   Lenvatinib. These data suggest AKT as an important mediator to
   Lenvatinib resistance.

Figure 5.

   [121]Figure 5
   [122]Open in a new tab

   Inhibition of the AKT pathway reverse Lenvatinib resistance in HCC. (A)
   Western blotting analysis of AKT and p-AKT expression in SNU-739 and
   Hepa1-6 cells transfected with siNC and siAKT. (B) Dose-response curves
   of SNU-739 and Hepa1-6 cells treated with Lenvatinib after transfection
   with either siNC or siAKT. (C) The growth curves show the fold change
   of SNU-739 and Hepa1-6 with siNC or siAKT treated with Lenvatinib (2.5
   μM). (D) Colony formation assay in SNU-739 and Hepa1-6 cells treated
   with Lenvatinib (2.5 μM) after knockdown AKT. (E) The quantification of
   the three independent assays shown in (D).

AKT inhibitor, AZD5363, sensitizes HCC cells to Lenvatinib

   Based on the above results that ATK depletion attenuated HCC cell
   resistance to Lenvatinib, we further employed a well-validated AKT
   inhibitor, AZD5363, to further confirm whether AKT inhibition
   contributes to overcome Lenvatinib resistance. It did show that
   combination of AZD5363 and Lenvatinib was more effective in suppressing
   cell survival than each single-drug, indicating a synergistic effect
   between AZD5363 and Lenvatinib (Fig. [123]6A-[124]6C). Additionally, we
   performed cell apoptosis assays, which revealed that the combined
   treatment significantly promoted apoptosis in SNU-739 and Hepa1-6
   cells, accompanied by an increased expression of cleaved PARP (Fig.
   [125]6D and S6A). Using syngeneic graft mouse model, we evaluated the
   combination effects of the two drugs in vivo. Compared to the groups
   treated with AZD5363 or Lenvatinib alone, the combined application of
   both drugs resulted in a significant reduction in tumor volume (Fig.
   [126]6E-[127]6G), without reducing mouse weight compared to vehicle
   control ([128]Fig. S6B). Immunohistochemistry results showed a
   significantly decreased Ki67 and increased cleaved caspase 3 positive
   rates in the combination group (Fig. [129]6H and [130]6I).
   Collectively, these data suggest that the AKT inhibitor AZD5363, in
   combination with Lenvatinib, exhibits synergistic effects and can
   reverse Lenvatinib resistance in liver cancer cells.

Figure 6.

   [131]Figure 6
   [132]Open in a new tab

   The AKT inhibitor AZD5363 synergizes with Lenvatinib in HCC. (A) Colony
   formation of SNU-739 and Hepa1-6 cells treated with either DMSO,
   Lenvatinib (2.5 μM), AZD5363 (1.25 μM) or their combination. (B) The
   quantification of the three independent assays shown in (A). (C) Growth
   curve of SNU-739 and Hepa1-6 treated with either DMSO, Lenvatinib (2.5
   μM), AZD5363 (1.25 μM) or combination. (D) Cell apoptosis analysis of
   SNU-739 and Hepa1-6 cells treated with either DMSO, Lenvatinib, AZD or
   combination. (E) Tumor growth curve of Hepa1-6 xenograft models. Mice
   were treated with vehicle, Lenvatinib (4 mg/kg), AZD5363 (100 mg/kg),
   or combination daily for 18 days. Growth curve was plotted by measuring
   the relative tumor volume every 3 days. (F-G) The excised tumors and
   relative tumor weight at the termination day from the experiment
   described in E, respectively. (H) Representative images of Ki-67 and
   cleaved caspase-3 staining in Hepa1-6 xenograft models. Scale bars, 25
   μm. (I) Quantification of Ki67 or cleaved caspase-3 positive rate of
   tumor from mice in each group.

   In order to explore the clinical relevance of combinatorial application
   of SAHA and Lenvatinib, we tested the combination effects of these two
   drugs in liver cancer organoids. We performed dose-response tests on 5
   primary liver cancer organoids with Lenvatinib (Figure [133]7A-B),
   which showed 2 of them (with IC[50]s less than 5 μM) were relatively
   sensitive to Lenvatinib treatment, while 3 of them were resistant to
   Lenvatinib (with IC[50]s greater than 20 μM). Interestingly, the p-AKT
   level were much higher in Lenvatinib-resistant organoids compared to
   sensitive ones (Figure [134]7C). Moreover, both the combinatorial
   application of SAHA/Lenvatinib, and AZD5363/Lenvatinib, exhibited
   extraordinarily better inhibitory effects than Lenvatinib alone in
   Lenvatinib-resistant tumor organoids, but not in Lenvatinib-sensitive
   ones (Fig. [135]7D and S7). These results support our hypothesis that
   Lenvatinib exerts anticancer effects partially by inhibiting MEK/ERK
   signaling pathway via targeting receptor tyrosine kinases, while AKT
   signaling activation attenuates HCC cells' sensitivity to Lenvatinib.
   Thus, targeting AKT by either HDACi or AKT inhibitor showed significant
   efficacy in Lenvatinib-resistant HCC cells and organoid, which may
   benefit HCC patients with AKT activation (Fig. [136]7E).

Figure 7.

   [137]Figure 7
   [138]Open in a new tab

   The synergistic effects of Lenvatinib with SAHA or AZD5363 were
   validated in HCC organoid models. (A) Dose-response curve of HCC
   organoids treated with Lenvatinib. (B) Average IC[50] values generated
   from dose-response curves from (A). (C) Western blotting analysis of
   the expression of total AKT and p-AKT in HCC organoids. (D) Synergy map
   of HCC organoids treated with the combination of Lenvatinib and SAHA or
   AZD5363. (E) Schematic models of the mechanism underlying SAHA or AKT
   inhibitor reverses Lenvatinib resistance in HCC. Lenvatinib mainly
   reduced the MAPK pathway signaling while activation of the AKT pathway
   allows cells to continue to proliferate and survive independent of MAPK
   signaling, blocking AKT signaling pathway by either HDACi or AKT
   inhibitor resensitized Lenvatinib resistant HCC to Lenvatinib
   treatment.

Discussion

   HCC is considered as one of the most challenging malignancies owing to
   its high incidence and mortality rates [139]^1. Lenvatinib has been
   approved as a first-line treatment in 2018 for unresectable or advanced
   HCC due to its non-inferior cure rate and improvement in
   progression-free survival compared to sorafenib [140]^4. Nevertheless,
   a significant proportion of HCC patients either exhibit no response to
   Lenvatinib or develop resistance during drug treatment, thereby
   diminishing its efficacy [141]^34. Hence, a comprehensive understanding
   of the molecular mechanisms underlying Lenvatinib resistance, the
   exploration of effective therapeutic strategies, are of utmost
   importance to improve treatment response for HCC patients.

   Lenvatinib is a well-defined tyrosine kinase inhibitor for multiple
   targets, including receptors VEGFR1-3, FGFR1-4, PDGFR, KIT, and RET
   [142]^6. These targets are well-demonstrated to promote cell growth,
   proliferation, and angiogenesis through activating various signaling,
   including MAPK and PI3K/AKT pathways, in HCC [143]^35. Numerous studies
   have determined various factors associated with Lenvatinib resistance,
   such as Jin et al. found that inhibition of receptor tyrosine kinases
   by Lenvatinib leads to feedback activation of EGFR-PAK2-ERK5 pathways,
   which reduced the effectiveness of Lenvatinib in HCC [144]^36. Besides,
   other mechanisms involved in cell metabolism [145]^37, epigenetics
   [146]^38^-[147]^40, autophagy [148]^41^, [149]^42, EMT [150]^43,
   noncoding RNAs [151]^44^-[152]^47, and the tumor microenvironment
   [153]^48, have also been suggested to be related to HCC Lenvatinib
   resistance.

   Several HDAC members, including HDAC1, HDAC2, HDAC4, HDAC7, etc., have
   been reported to correlated with HCC progression and patient survival
   [154]^49, and HDAC2 expression is upregulated in Lenvatinib-resistant
   cells [155]^50, suggesting a potential role of HDACs in Lenvatinib
   resistance. Accordingly, through a pharmacological screen of potential
   compounds that might conquer Lenvatinib-resistance in liver cancer, we
   identified SAHA as a candidate using Lenvatinib-resistant organoids and
   confirmed this result in more cell lines and organoids. HDAC
   inhibitors, like other epigenetic-based therapeutic strategies, induce
   extensive transcriptional alterations, and subsequent growth arrest,
   differentiation inhibition and cell death in cancer cells, thus exhibit
   anti-tumor efficacy in a variety of cancer types [156]^11. Although
   HDACis have shown very positive pre-clinical results and four compounds
   have received clinical approval, only limited success was achieved when
   used as mono-therapeutic agents against solid tumors in clinical trials
   compared to the hematological malignancies [157]^51. Meanwhile, side
   effects and toxicities from HDACis also hinder their progress in the
   clinic [158]^52^, [159]^53. Thus, administering HDACis in low doses
   and, meanwhile, in combination with appropriate compounds might
   mitigate the drug toxicity and optimize curative effects
   simultaneously. Therefore, current clinical trials frequently combine
   HDAC inhibitors with chemotherapy or other targeted therapies to
   enhance clinical efficacy. We also noticed in our study that
   application of low dose SAHA did not obviously suppress HCC cell
   viability in mouse and organoid models (Figs. [160]2H and S7), while
   combinatorial use of low dose SAHA and Lenvatinib severely reduced cell
   viabilities.

   We further revealed PTEN/PI3K/AKT pathway as the major signaling that
   was down-regulated by HDACi in Lenvatinib-resistant cells, suggesting
   activation of such pathway contributes to Lenvatinib resistance of HCC.
   The PI3K/AKT signaling pathway is considered to be one of the most
   crucial pathways involved in cancer initiation and progression
   [161]^23, and several studies have underlined a potential role of
   PI3K/AKT activation in promoting drug resistance of HCC [162]^29^,
   [163]^54. Indeed, we found ectopic overexpression of a constitutive
   active form of AKT in Lenvatinib sensitive HCC cells promotes cell
   resistance to Lenvatinib treatment (Fig. [164]4), while depletion of
   AKT in Lenvatinib resistant HCC cells restored sensitivity to
   Lenvatinib treatment (Fig. [165]5). Even so, we also noticed HDACi
   achieved better tumor-suppression efficacy than AKT inhibition in both
   in vitro and in vivo, in both single and combinatorial treatment (Fig.
   [166]2 and Fig. [167]6), suggesting HDACis might also function through
   inhibiting signaling pathways other than PI3K/AKT pathway, such as MAPK
   signaling pathway (Fig. [168]3B).

   Taken together, we identified HDACi SAHA as a potential resolution for
   overcoming HCC Lenvatinib resistance via an HCC organoid-based drug
   screening in this study. Further mechanism research revealed
   PTEN/PI3K/AKT pathway as the major signaling in mediating
   Lenvatinib-resistant. Our data indicated that a subset of HCC cell
   lines or organoids displaying high levels of AKT phosphorylation
   exhibits resistance to Lenvatinib. More importantly, either HDACi or
   AKTi could sensitize these HCC cells to Lenvatinib, the combination of
   Lenvatinib with SAHA or AZD5363 targets the AKT signaling pathway could
   overcome Lenvatinib resistance. Our results offer substantial
   experimental evidences and comprehensive mechanistic investigation to
   support a novel combination therapy strategy for the treatment of
   advanced HCC.

Supplementary Material

   Supplementary figures and tables.
   [169]ijbsv20p3046s1.pdf^ (3.2MB, pdf)

Acknowledgments