Abstract

Background & Aims

   Lysosome-associated protein transmembrane 4β (LAPTM4B) is an oncogene
   implicated in the malignant progression of hepatocellular carcinoma
   (HCC). Previous research established a strong association between
   LAPTM4B and HCC stemness. However, specific mechanisms by which LAPTM4B
   regulates and maintains the stemness of liver cancer stem cells remain
   unclear. Therefore, we investigated the effects of LAPTM4B on the
   stemness regulation of cluster of differentiation 133 (CD133)^+ liver
   cancer stem-like cells (CSLCs).

Methods

   We used RNA interference and overexpression techniques in both in vitro
   and in vivo models. The involvement of LAPTM4B in wingless/integrated
   (WNT)/β-catenin signaling was examined through western blotting,
   immunofluorescence, and immunoprecipitation. The impact of LAPTM4B on
   β-catenin phosphorylation and ubiquitination was analyzed to elucidate
   its role in promoting stemness. Clinical relevance was evaluated in an
   in-house cohort of 105 specimens from patients with HCC through
   immunohistochemical and microarray analysis, enabling investigation of
   correlations with clinical outcomes.

Results

   LAPTM4B promoted the self-renewal ability, chemoresistance, and
   tumorigenicity of CD133^+ CSLCs. Mechanistically, aberrant LAPTM4B
   upregulation facilitated β-catenin nuclear translocation
   (nucleocytoplasmic separation assay, p <0.001) and inhibited its
   phosphorylation (p <0.01). In addition, LAPTM4B interacts with the
   deubiquitinating enzymes ubiquitin carboxyl-terminal hydrolase (USP)-1
   and USP14, reducing β-catenin ubiquitination. Furthermore, patients
   with high LAPTM4B and β-catenin expression had markedly shorter 3-year
   overall survival rate (42.9% vs. 74.4%; hazard ratio, 5.174; 95% CI
   2.280–11.741, p <0.001).

Conclusions

   LAPTM4B promotes CD133^+ CSLC stemness by activating WNT/β-catenin
   signaling by inhibiting β-catenin phosphorylation and ubiquitination
   degradation. The role of LAPTM4B in regulating WNT/β-catenin signaling
   suggests that LAPTM4B serves as a therapeutic target for impairing HCC
   stemness and progression.

Impact and implications

   LAPTM4B contributes significantly to CD133^+ CSLC stemness and inhibits
   β-catenin phosphorylation and ubiquitination degradation, activating
   WNT/β-catenin signaling. WNT inhibitors suppress LAPTM4B-induced
   CD133^+ CSLC stemness. Thus, targeting the LAPTM4B–WNT/β-catenin axis
   could improve antitumor efficacy.

   Keywords: LAPTM4B, β-Catenin stabilization, Ubiquitination,
   Deubiquitinating enzymes, WNT signaling inhibitors, Cancer stem cells,
   Hepatocellular carcinoma

Graphical abstract

   [31]Image 1
   [32]Open in a new tab

Highlights

     * •
       LAPTM4B enhances the stemness of CD133^+ liver CSLCs.
     * •
       LAPTM4B promotes CD133^+ CSLC stemness by activating WNT/β-catenin
       signaling.
     * •
       LAPTM4B inhibits β-catenin phosphorylation and ubiquitin-mediated
       degradation.
     * •
       High levels of LAPTM4B and β-catenin predict poor prognoses in
       patients with HCC.

Introduction

   Hepatocellular carcinoma (HCC) is the sixth most frequent malignant
   tumor worldwide, posing a significant threat to human life.[33]^1 Liver
   cancer stem cells, a distinct subset of HCC cells with stem cell
   features, are a fundamental cause of tumor heterogeneity and contribute
   to metastasis, recurrence, and drug resistance.[34]^2^,[35]^3

   Lysosome-associated protein transmembrane-4β (LAPTM4B) is markedly
   upregulated in poorly differentiated HCC tissues.[36]^4 It primarily
   mediates cell signal transduction, intracellular protein localization
   and isolation,[37]^5 and protein degradation, especially ubiquitin
   degradation.[38]^6 Overexpression of LAPTM4B can result in uncontrolled
   cell proliferation and malignant transformation, suggesting its role as
   an oncogene in HCC development.[39]^7 Our previous single cell
   RNA-sequencing (RNA-seq) study identified an association between
   LAPTM4B and HCC stemness.[40]^8 We sorted cluster of differentiation
   133 (CD133)-positive liver cancer stem-like cells (CSLCs), which are
   highly heterogeneous and exhibit stem cells traits, from HCC cells.
   CD133, a critical surface protein marker, is widely recognized for its
   functional and phenotypic significance in liver cancer stem
   cells.[41]^9^,[42]^10 Subsequently, we further explored the functional
   role and regulatory mechanism of LAPTM4B in CD133^+ CSLCs.

   Cancer stem cells maintain their stemness through multiple pathways,
   with wingless/integrated (WNT)/β-catenin signaling as the most commonly
   activated pathway in HCC.[43]^11 With evolutionarily high conservatism,
   the WNT/β-catenin pathway supports tumor growth, metastasis,
   recurrence, and radiochemotherapy resistance.[44][12], [45][13],
   [46][14] The destruction complex comprising adenomatous polyposis coli
   (APC), axis inhibition protein, casein kinase 1, and glycogen synthase
   kinase 3β (GSK-3β) regulates β-catenin levels and nuclear
   translocation.[47]^15 When the WNT/β-catenin pathway is activated,
   β-catenin is released from the destruction complex, and avoids being
   phosphorylated and targeted for ubiquitin-mediated proteasomal
   degradation. It then accumulates in the cytoplasm before moving into
   the nucleus to trigger the transcription and expression of target
   genes, such as those encoding cellular myelocytomatosis (c-MYC), Cyclin
   D1, and matrix metalloproteinase 7 (MMP7).[48][16], [49][17], [50][18],
   [51][19]

   Inhibiting the stemness of HCC cells could provide new therapeutic
   strategies for HCC treatment. However, the mechanisms that regulate and
   maintain stem cell characteristics remain unclear. Therefore, an
   in-depth exploration of these molecular mechanisms and their regulatory
   networks is essential for improving HCC treatment and prognosis. Thus,
   we investigated the effect of LAPTM4B on the stemness regulation of
   CD133^+ CSLCs. These findings provide potential insight into
   LAPTM4B-mediated stemness, the mechanisms of action of the
   WNT/β-catenin pathway, and its function in cancer progression.

Material and methods

   We refer readers to the supplementary data for further details of: HCC
   cell lines and culture; plasmids, lentiviruses, and experimental
   protocol for cell transfection; isolation of CD133^+ CSLCs; microsphere
   cultures and sphere-forming assays; RNA-seq; immunofluorescence (IF)
   staining; construction of the xenograft tumor model; and selection of
   human specimens and immunohistochemical (IHC) staining.

Co-immunoprecipitation assays

   According to the manufacturer’s (Beyotime Biotech, Shanghai, China)
   instructions, cells were lysed with lysis buffer, and the whole-cell
   extracts were supplemented with the protease inhibitor PMSF. The
   lysates were then incubated with 2–5 ug IP antibodies ([52]Table S5)
   overnight at 4 °C on a rotating wheel. IgG was used as the negative
   control in each experiment. The complexes were incubated with protein
   A/G magnetic beads (HY-K0202, MedChemExpress) at 4 °C. The precipitates
   were washed six times with ice-cold wash buffer, eluted with 0.1 M
   glycine-HCl (pH 3.0) buffer, and resolved by SDS-PAGE followed by
   western blot analysis with appropriate antibodies ([53]Table S5).

Protein stability experiment

   To measure the half-life of β-catenin, cells were treated with
   100 μg/ml protein synthesis inhibitor cycloheximide (HY12320;
   MedChemExpress) for 0 h, 1 h, 4 h, 8 h, 12 h, and 24 h. Western
   blotting was performed to measure protein levels. LAPTM4B-depleted or
   -overexpressing cells and control cells were incubated with 10 μM MG132
   (HY13259; MedChemExpress) for 12 h. Following this treatment, total
   protein was extracted for western blot analysis.

Ubiquitination assay

   To detect ubiquitination of β-catenin, cells were incubated with MG132
   (10 μM) for 8 h and then lysed by RIPA buffer. Proteins in the cell
   lysate were immunoprecipitated to isolate ubiquitinated β-catenin with
   anti-β-catenin antibodies and the ubiquitination level of β-catenin
   were detected through western blotting with antibodies against
   ubiquitin.

Statistical analysis

   Statistical analysis was conducted using SPSS software version 22.0
   (SPSS, Chicago, IL, USA). Data are given as mean ± SD. The Student’s t
   test was used to assess differences between groups, and one-way ANOVA
   was used to assess variations between multiple groups. Clinical data
   were examined using Fisher’s exact test. The Kaplan-Meier method was
   used to determine survival rates, with significance assessed using the
   Log-rank test. Pearson correlation was used to assess the relationship
   between gene expression levels. Regression analysis was estimated using
   the Cox proportional hazards model after confirmation of the
   proportional hazard assumption and described by hazard ratios (HR) and
   95% CIs. p<0.05 indicated statistical significance.

Results

LAPTM4B regulates stemness in CD133^+ CSLCs

   Previous research demonstrated that LAPTM4B enhances invasion,
   metastasis, and tumor initiation in HCC.[54]^8^,[55]^20 Here, we
   established LAPTM4B-overexpressing Huh7 and HepG2 cells, achieving over
   a threefold increase in LAPTM4B expression ([56]Fig. S1A,B).
   Intriguingly, IF analysis indicated that LAPTM4B overexpression
   increased the proportion of CD133-positive cells (a key stem cell
   marker in HCC[57]^21^,[58]^22) ([59]Fig. S1C–F), as also evidenced by
   flow cytometry ([60]Fig. S1G,H). In addition, LAPTM4B expression was
   higher in CD133^+ cells than in CD133^– cells ([61]Fig. S1I,J). IHC
   analysis of 105 HCC specimens with long-term clinical follow-up data
   revealed a positive association between LAPTM4B levels and CD133
   expression ([62]Fig. S1K,L). These findings prompted us to investigate
   the role of LAPTM4B in regulating CD133^+ CSLC stemness. Using
   immunomagnetic sorting, we isolated CD133^+ CSLCs from Huh7 and HepG2
   cells, achieving high purity confirmed by flow cytometry
   ([63]Fig. S2A,B). Sphere-formation assays demonstrated that isolated
   CD133^+ CSLCs from both cell lines displayed excellent sphere-forming
   capacity ([64]Fig. S2C–F). Western blotting revealed that expression of
   CD133, EpCAM, SOX9, OCT4, and NANOG was higher in CD133^+ CSLCs than in
   CD133^– tumor cells ([65]Fig. S2G,H). These results indicated a
   stem-like phenotype in CD133^+ CSLCs isolated from HCC cells.

   Subsequently, we intervened in LAPTM4B expression in CD133^+ CSLCs and
   observed that LAPTM4B knockdown reduced expression by over 60%
   ([66]Fig. S3A,B). Reduced sphere-forming ability of CD133^+ CSLCs was
   also observed with LAPTM4B knockdown ([67]Fig. 1A,B). Conversely,
   overexpression enhanced sphere formation ([68]Fig. S4A–D). LAPTM4B
   suppression significantly reduced several genes typically associated
   with stemness, including SOX9, OCT4, and NANOG, as well as a stem cell
   surface marker, EpCAM ([69]Fig. 1C–F), whereas overexpression increased
   these markers ([70]Fig. S4E–H).

Fig. 1.

   [71]Fig. 1
   [72]Open in a new tab

   LAPTM4B enhances CD133^+ CSLC stemness.

   (A,B) Effect of LAPTM4B-knockdown on sphere-forming ability of CD133^+
   Huh7 and CD133^+ HepG2 cells on day 5. (C–F) qPCR and western blot
   analysis of EpCAM, SOX9, OCT4 and NANOG expression in CD133^+ Huh7 and
   CD133^+ HepG2 cells after LAPTM4B knockdown. (G,H) Effect of LAPTM4B
   knockdown on EMT in CD133^+ Huh7 and CD133^+ HepG2 cells analyzed by
   western blotting. (I) Effect of silencing LAPTM4B on CD133^+ Huh7 and
   CD133^+ HepG2 cells apoptosis in presence of lenvatinib (10 μM, 72 h).
   (J) Survival curves of CD133^+ Huh7/CD133^+ HepG2
   LAPTM4B-overexpressing and control cells treated with lenvatinib for
   72 h. (K,L) Colony formation assay to detect CD133^+ Huh7/CD133^+ HepG2
   proliferation with LAPTM4B knockdown or overexpression in presence of
   lenvatinib. (M,N) Subcutaneous xenograft experiment evaluating
   tumor-formation capability of LAPTM4B-knockdown CD133^+ Huh7 cells (n =
   7 per group). (O) Immunohistochemical staining of LAPTM4B, EpCAM, and
   SOX9 in LAPTM4B-knockdown and control cells from xenografted tumors.
   Data are means ± SD for at least 2 independent experiments, analyzed
   with unpaired Student’s t tests: ∗p <0.05; ∗∗p <0.01; ∗∗∗p <0.001.
   Scale bars: 200 μm (A,B), 100 μm (O), 1.5 cm (M,N). See also [73]Figs
   S1–S4. CD133, cluster of differentiation 133; CSLC, cancer stem-like
   cell; EMT, epithelial–mesenchymal transition; LAPTM4B,
   lysosome-associated protein transmembrane-4β; qPCR, quantitative
   real-time PCR.

   Given that cancer stem cells often exhibit drug resistance and evade
   treatments through mechanisms such as the epithelial-to-mesenchymal
   transition (EMT),[74]^23^,[75]^24 we investigated EMT markers in
   LAPTM4B-altered CSLCs. LAPTM4B knockdown upregulated the epithelial
   marker E-cadherin and reduced the mesenchymal markers N-cadherin,
   vimentin, and snail2 ([76]Fig. 1G,H), whereas overexpressed LAPTM4B
   induced EMT in CD133^+ CSLCs ([77]Fig. S4I,J). To determine the impact
   of LAPTM4B on the drug resistance of CD133^+ CSLCs, we examined the
   apoptosis and proliferation of CD133^+ CSLCs in the presence of
   lenvatinib. LAPTM4B knockdown increased sensitivity to
   lenvatinib-induced apoptosis ([78]Fig. 1I; [79]Fig. S3C), whereas
   overexpression conferred resistance ([80]Fig. S4K,L) and protected
   against lenvatinib-induced growth inhibition ([81]Fig. 1J–L).
   Furthermore, we explored whether LAPTM4B promotes the tumor-forming
   ability of cells by injecting CD133^+ CSLCs subcutaneously with or
   without LAPTM4B knockdown in nude mice. The tumor volume of mice
   injected with LAPTM4B-knockdown CD133^+ Huh7 cells was significantly
   reduced compared with that in the control group ([82]Fig. 1M,N).
   Limiting dilution assays further showed that LAPTM4B overexpression
   enhanced tumor formation in vivo ([83]Fig. S4M). Correspondingly, IHC
   staining of the tumor tissues revealed lower expression of stemness
   markers in LAPTM4B-knockdown tumors ([84]Fig. 1O). The above results
   revealed that LAPTM4B has an essential role in promoting CD133^+ CSLC
   stemness in HCC.

LAPTM4B activates the WNT/β-catenin pathway in CD133^+ CSLCs

   We analyzed the expression profiles of LAPTM4B-silence or control
   CD133^+ Huh7 cells to gain a broader understanding of the changes in
   LAPTM4B-mediated genes and their effects on the CD133^+ CSLC phenotype.
   Kyoto Encyclopedia of Genes and Genomes pathway enrichment and Gene
   Ontology functional category enrichment analyses showed that the
   regulation of CD133^+ Huh7 cell stemness by LAPTM4B involved multiple
   biological processes, including G2/M transition of the mitotic cell
   cycle, positive regulation related to cell morphology differentiation,
   WNT signaling pathway, and protein dephosphorylation ([85]Fig. 2A).
   Given that hyperactivation of WNT/β-catenin signaling significantly
   contributes to liver cancer stem cell stemness,[86]^25^,[87]^26 we
   focused on the regulation of WNT/β-catenin signaling by LAPTM4B. We
   observed increased mRNA levels of GSK-3β and APC in CD133^+ Huh7 and
   CD133^+ HepG2 cells following LAPTM4B knockdown, whereas c-Myc and
   Cyclin D1 decreased ([88]Fig. 2B,C). The protein level of GSK-3β was
   elevated, whereas the expression of p-GSK-3β, c-Myc, and Cyclin D1
   declined when LAPTM4B was silenced ([89]Fig. 2D,E). Conversely,
   overexpressing LAPTM4B resulted in opposite effects ([90]Fig. S5A–D).
   Nuclear–cytoplasmic shuttling of β-catenin is a pivotal step in
   WNT/β-catenin pathway activation.[91]^27 As expected, nucleocytoplasmic
   separation assays and IF indicated enrichment of β-catenin nuclear
   localization in LAPTM4B-overexpressing CD133^+ CSLCs ([92]Fig. 2F–I)
   and also at the organizational level ([93]Fig. 2J,K). These
   observations demonstrated that LAPTM4B activated the WNT/β-catenin
   pathway in CD133^+ CSLCs.

Fig. 2.

   [94]Fig. 2
   [95]Open in a new tab

   LAPTM4B affects WNT/β-catenin pathway activation.

   (A) Pathway enrichment analysis in LAPTM4B-knockdown CD133^+ Huh7 or
   control cells identified by RNA sequencing. (B,C) GSK-3β, APC, c-Myc
   and Cyclin D1 expression in CD133^+ Huh7 and CD133^+ HepG2 cells with
   LAPTM4B knockdown. (D,E) Western blots of GSK-3β, p-GSK-3β, c-Myc, and
   Cyclin D1 expression in CD133^+ Huh7 and CD133^+ HepG2 cells when
   silencing LAPTM4B. (F) Western blots of nucleocytoplasmic distribution
   of β-catenin in LAPTM4B-overexpressing CD133^+ Huh7 cells. (G) Relative
   protein levels of nuclear β-catenin in CD133^+ Huh7 cells
   overexpressing LAPTM4B or not. (H,I) Immunofluorescence of nuclear
   translocation of β-catenin. (J,K) Colocalization of LAPTM4B and
   β-catenin in HCC examined by immunofluorescence and
   immunohistochemistry. Data are means ± SD for at least 2 independent
   experiments, analyzed with unpaired Student’s t tests: ∗p <0.05; ∗∗p
   <0.01; ∗∗∗p <0.001. Scale bars: 100 μm (J,K); 25 μm (H,I). See also
   [96]Fig. S5. APC, adenomatous polyposis coli; c-MYC, cellular
   myelocytomatosis; CD133, cluster of differentiation 133; GSK-3β,
   glycogen synthase 3β; HCC, hepatocellular carcinoma; LAPTM4B,
   lysosome-associated protein transmembrane-4β; p-GSK-3β, phosphorylated
   glycogen synthase 3β; qPCR, quantitative real-time PCR.

LAPTM4B stabilizes β-catenin by inhibiting its degradation

   Phosphorylated β-catenin (Thr41/Ser45) undergoes ubiquitination
   degradation, resulting in a closed WNT/β-catenin pathway.[97]^28 We
   observed that β-catenin decreased at both the mRNA and protein levels
   in LAPTM4B-deficient CD133^+ CSLCs, whereas the phosphorylated
   β-catenin (Thr41/Ser45) protein level increased ([98]Fig. 3A–D). The
   opposite results were achieved by overexpressing LAPTM4B
   ([99]Fig. S5E–H). This result validated the robustness of the RNA-seq
   results ([100]Fig. 2A) and further suggested that LAPTM4B promotes WNT
   signaling activation by modulating phosphorylation of β-catenin.

Fig. 3.

   [101]Fig. 3
   [102]Open in a new tab

   Effect of LAPTM4B on β-catenin degradation.

   (A,B) qPCR detection of β-catenin expression in CD133^+ Huh7 and
   CD133^+ HepG2 cells following silencing of LAPTM4B. (C,D) Western blots
   of β-catenin and p-β-catenin expression in CD133^+ Huh7 and CD133^+
   HepG2 cells following LAPTM4B knockdown. (E) β-catenin and p-β-catenin
   protein levels in LAPTM4B-knockdown CD133^+ Huh7 cells after CHX
   treatment. (F) Half-life of β-catenin in LAPTM4B-knockdown CD133^+ Huh7
   cells after CHX treatment. (G) Western blots of β-catenin levels in
   CD133^+ Huh7 cells with LAPTM4B knockdown after treatment with CHX and
   MG132. (H) Western blot in vitro ubiquitination assay of β-catenin in
   CD133^+ Huh7 cells with LAPTM4B knockdown after MG132 treatment. (I)
   β-catenin and p-β-catenin levels in LAPTM4B-overexpressing CD133^+
   HepG2 cells after CHX treatment. (J) Half-life of β-catenin in
   LAPTM4B-overexpressing CD133^+ HepG2 cells after CHX treatment. (K)
   Western blots of β-catenin levels in LAPTM4B-overexpressing CD133^+
   HepG2 cells after CHX and MG132 treatment. (L) Western blot in vitro
   ubiquitination assay of β-catenin in LAPTM4B-overexpressing CD133^+
   HepG2 cells after MG132 treatment. Data are means ± SD for at least two
   independent experiments, analyzed with unpaired Student’s t tests: ∗p
   <0.05; ∗∗p <0.01; ∗∗∗p <0.001. CD133, cluster of differentiation 133;
   CHX, cycloheximide; LAPTM4B, lysosome-associated protein
   transmembrane-4β; qPCR, quantitative real-time PCR.

   Previous studies have shown that LAPTM4B inhibits ubiquitination, which
   prevents protein degradation,[103]^6^,[104]^8 and that ubiquitinated
   degradation causes poor stability of β-catenin in the
   cytoplasm.[105]^15 Thus, we hypothesized that LAPTM4B regulates
   β-catenin protein stability by affecting its ubiquitinated degradation.
   Cycloheximide (CHX) chase assays were conducted to identify the
   underlying molecular mechanism of the role of LAPTM4B in β-catenin
   stability. With increased CHX treatment time, the degradation of
   β-catenin increased and the half-life of β-catenin decreased in
   LAPTM4B-depleted cells ([106]Fig. 3E,F). In addition, LAPTM4B knockdown
   attenuated the inhibitory effects of the proteasome inhibitor MG132 on
   β-catenin degradation ([107]Fig. 3G). MG132 was administered to CD133^+
   Huh7 cells to further validate the role of LAPTM4B in inhibiting the
   ubiquitination and subsequent degradation of β-catenin. Ubiquitination
   assays demonstrated that LAPTM4B knockdown increased ubiquitinated
   β-catenin in CD133^+ Huh7 cells ([108]Fig. 3H). By contrast, LAPTM4B
   overexpression prolonged the half-life of β-catenin in CD133^+ HepG2
   cells ([109]Fig. 3I,J). Furthermore, LAPTM4B overexpression enhanced
   the inhibitory effect of MG132 on β-catenin degradation ([110]Fig. 3K).
   The ubiquitination assays confirmed that β-catenin protein
   ubiquitination was significantly reduced in LAPTM4B-overexpressing
   cells ([111]Fig. 3L). These outcomes showed that LAPTM4B promotes
   β-catenin stability by inhibiting its ubiquitin–proteasome pathway
   degradation.

LAPTM4B inhibits β-catenin degradation by binding to USP1 and USP14

   We demonstrated that LAPTM4B promotes β-catenin stability by inhibiting
   its ubiquitination. Here we used three side-by-side datasets: the
   deubiquitinating enzyme dataset, the gene set co-expressed with LAPTM4B
   and β-catenin from the TCGA database, and the gene set interacting with
   CTNNB1 from the STRING database. We found that the deubiquitinating
   enzymes ubiquitin carboxyl-terminal hydrolase (USP)-1 and USP14 are
   involved in the inhibition of β-catenin ubiquitination degradation by
   LAPTM4B ([112]Fig. 4A). The Gene Expression Profiling Interactive
   Analysis (GEPIA) platform online analysis identified a significant
   positive correlation between LAPTM4B expression and both USP1 and USP14
   in HCC ([113]Fig. 4B). The UbiBrowser database highlighted USP1 as a
   predicted and USP14 as a known deubiquitinating enzyme for β-catenin
   ([114]Fig. 4C). Co-immunoprecipitation (Co-IP) assays confirmed the
   interaction between LAPTM4B and USP1 ([115]Fig. 4D). In ubiquitination
   assays, LAPTM4B overexpression decreased β-catenin ubiquitination,
   whereas knockdown of USP1 reversed the inhibitory effect of LAPTM4B on
   β-catenin ubiquitinated-degradation ([116]Fig. 4E). Similar results
   were observed between LAPTM4B and USP14 ([117]Fig. 4F,G). Meanwhile,
   Co-IP assays using anti-β-catenin antibodies reconfirmed that LAPTM4B
   interacted with β-catenin, USP1, and USP14.

Fig. 4.

   [118]Fig. 4
   [119]Open in a new tab

   LAPTM4B and USP1/USP14 interaction inhibit β-catenin ubiquitinated
   degradation.

   (A) Comprehensive analysis combining the deubiquitinating enzyme
   dataset, genes co-expressed with LAPTM4B/β-catenin from the TCGA
   database, and genes interacting with CTNNB1 in the STRING database. (B)
   Pearson’s correlation analysis of the association between LAPTM4B and
   USP1/USP14 expression in patients with HCC from the TCGA database. (C)
   Predicted and known deubiquitinating enzymes of β-catenin from the
   UbiBrowser database. (D) Co-IP of the interaction between LAPTM4B and
   USP1. (E) β-catenin in CD133^+ Huh7 cells with LAPTM4B-overexpression
   and USP1-knockdown after MG132 treatment used western blotting in vitro
   ubiquitination assay. (F) Co-IP of the interaction between LAPTM4B and
   USP14. (G) Western blot in vitro ubiquitination assay of β-catenin in
   CD133^+ Huh7 cells with LAPTM4B overexpression and USP14 knockdown
   after MG132 treatment. (H) Schematic of LAPTM4B protein truncation. (I)
   Co-IP assay using anti-Flag antibodies with truncated LAPTM4B
   overexpression. See also [120]Fig. S6. Co-IP, co-immunoprecipitation;
   LAPTM4B, lysosome-associated protein transmembrane-4β; STRING, Search
   Tool for the Retrieval of Interacting Genes/Proteins; TCGA, The Cancer
   Genome Atlas; USP, ubiquitin carboxyl-terminal hydrolase.

   Given that LAPTM4B might inhibit β-catenin ubiquitination degradation
   by assisting USP1 and USP14 in binding to β-catenin ([121]Fig. S6A), we
   overexpressed various LAPTM4B truncations in CD133^+ Huh7 cells to
   further clarify the interaction domain of LAPTM4B with USP1 and USP14.
   The interaction was disrupted only in cells overexpressing the
   FlagΔ265-317-LAPTM4B truncation ([122]Fig. 4H,I). These findings
   suggest that LAPTM4B inhibits β-catenin ubiquitination and stabilizes
   it through interactions with USP1 or USP14 via its 265–317 structural
   domain.

LAPTM4B promotes the stemness of CD133^+ CSLCs via WNT/β-catenin signaling

   The above results indicated that LAPTM4B enhances the stability of
   β-catenin by suppressing its phosphorylation and
   ubiquitination-mediated degradation. Therefore, we investigated the
   crucial involvement of WNT/β-catenin signaling in the modulation of
   CD133^+ CSLC stemness by LAPTM4B. Sphere-forming assays showed that
   LAPTM4B deficiency reduced the formation of spheres in CD133^+ Huh7
   cells, which was partially rescued by β-catenin overexpression
   ([123]Fig. 5A,B). Conversely, LAPTM4B upregulation facilitated
   the forming of spheres, whereas β-catenin suppression
   prevented LAPTM4B-induced sphere formation in CD133^+ Huh7 cells
   ([124]Fig. S7A,B). Similarly, the sphericity of CD133^+ HepG2 cells
   decreased after knocking down LAPTM4B, but improved when using LiCl (a
   WNT/β-catenin pathway activator[125]^29^,[126]^30) ([127]Fig. 5C,D).
   The expression of stemness-related genes, EpCAM, SOX9, OCT4, and NANOG
   decreased when LAPTM4B was silenced, whereas β-catenin overexpression
   reversed this effect in CD133^+ Huh7 cells ([128]Fig. 5E,F). However,
   overexpressing LAPTM4B increased the expression of these genes, which
   was suppressed by depleting β-catenin ([129]Fig. S7C,D). Likewise, LiCl
   prevented the downregulation of EpCAM, SOX9, OCT4, and NANOG caused by
   LAPTM4B knockdown in CD133^+ HepG2 cells ([130]Fig. 5G,H). Moreover,
   overexpression of LAPTM4B reduced apoptosis in response to lenvatinib
   treatment, whereas subsequent β-catenin knockdown significantly
   increased apoptosis in CD133^+ Huh7 and CD133^+ HepG2 cells
   ([131]Fig. 5I,J). Consistent with the above findings, LAPTM4B
   overexpression led to CD133^+ Huh7 cells resistant to
   lenvatinib-induced growth inhibition. However, this resistance was
   substantially diminished following β-catenin knockdown post-LAPTM4B
   overexpression ([132]Fig. 5K).

Fig. 5.

   [133]Fig. 5
   [134]Open in a new tab

   WNT/β-catenin pathway is responsible for stemness properties of
   LAPTM4B-mediated CD133^+ CSLCs.

   (A,B) Sphere-forming assay of CD133^+ Huh7 cells with LAPTM4B knockdown
   and overexpressing CTNNB1 in stem cell medium. (C,D) Sphere-forming
   assay of the effect of LiCl (WNT pathway activator) on the sphere
   formation ability of CD133^+ HepG2 cells with LAPTM4B knockdown. (E,F)
   Western blots of EpCAM, SOX9, OCT4, and NANOG expression in
   LAPTM4B-knockdown CD133^+ Huh7 cells with or without re-introduction of
   CTNNB1. (G,H) Western blots of EpCAM, SOX9, OCT4, and NANOG expression
   in LAPTM4B-knockdown CD133^+ HepG2 cells treated with LiCl. (I,J)
   Effect of overexpressing LAPTM4B and knocking down CTNNB1 on CD133^+
   Huh7 and CD133^+ HepG2 cell apoptosis in the presence of lenvatinib
   (20 μM, 72 h). (K) Colony growth of LAPTM4B-overexpressing CD133^+ Huh7
   cells with knocked-down CTNNB1 treated with lenvatinib for 7 days. (L)
   Tumor formation capacity of LAPTM4B-knockdown CD133^+ Huh7 cells with
   or without CTNNB1 overexpression (n = 7 per group). (M)
   Immunohistochemistry of β-catenin, EpCAM, and SOX9 expression in
   xenografted tumor tissue. Data are means ± SD for at least two
   independent experiments, analyzed with unpaired Student’s t tests: ∗p
   <0.05; ∗∗p <0.01; ∗∗∗p <0.001; n.s., no significant difference. Scale
   bars: 200 μm (A–D). 100 μm (M), 1.5 cm (L). See also [135]Fig. S7.
   CSLC, cancer stem-like cell; LAPTM4B, lysosome-associated protein
   transmembrane-4β; WNT, wingless/integrated.

   We further investigated the effects of WNT/β-catenin signaling on the
   regulation of HCC stemness by LAPTM4B in vivo. Mice were inoculated
   subcutaneously with CD133^+ Huh7-Vector, CD133^+ Huh7-shLAPTM4B, or
   CD133^+ Huh7-shLAPTM4B-CTNNB1 cells (5 × 10^6 cells per mouse). The
   tumor volume in the mice injected with the CD133^+ Huh7-shLAPTM4B cells
   was dramatically reduced compared with that in the vector group.
   However, restoration of β-catenin expression led to an increase in
   tumorigenicity ([136]Fig. 5L). IHC staining of the tumor tissues
   verified the correlation between LAPTM4B, β-catenin, and the levels of
   stemness-related genes in vivo ([137]Fig. 5M). In addition, limiting
   dilution subcutaneous xenografts indicated that LAPTM4B overexpression
   enhanced the capacity of tumor formation in CD133^+ Huh7 cells, whereas
   silencing β-catenin counteracted the tumor-promoting effects of LAPTM4B
   ([138]Fig. S7E). These observations indicated that LAPTM4B enhances the
   stemness of CD133^+ CSLCs through WNT/β-catenin signaling.

WNT inhibition suppresses LAPTM4B-induced CD133^+ CSLC stemness activation

   XAV-939 (a WNT/β-catenin signaling inhibitor[139][30], [140][31],
   [141][32]) was added during the suspension culture of CD133^+ CSLCs
   overexpressing LAPTM4B to preliminarily investigate the potential
   therapeutic effectiveness of targeting the WNT/β-catenin pathway in
   inhibiting LAPTM4B-induced HCC with high stemness characteristics. As
   previously described, LAPTM4B overexpression improved the
   sphere-forming ability of CD133^+ CSLCs, whereas XAV-939 attenuated it.
   The sphere-forming capacity of CD133^+ CSLCs overexpressing LAPTM4B was
   reduced but not significantly so after XAV-939 treatment
   ([142]Fig. 6A,B). The expression of β-catenin decreased under the
   action of XAV-939, whereas the addition of XAV-939 to the
   LAPTM4B-overexpressing group facilitated higher levels of β-catenin
   compared with the non-LAPTM4B-overexpressing group ([143]Fig. 6C,D).
   Moreover, the expression of the stemness-related genes EpCAM, SOX9,
   OCT4, and NANOG was downregulated under XAV-939, although
   overexpression of LAPTM4B prevented this inhibitory effect to some
   extent ([144]Fig. 6E–G). These results suggest that the WNT/β-catenin
   pathway is crucial for LAPTM4B-regulated stemness in CD133^+ CSLCs and
   lays the foundation for considering the LAPTM4B-WNT/β-catenin axis as a
   potential therapeutic target for the treatment of HCC.

Fig. 6.

   [145]Fig. 6
   [146]Open in a new tab

   LAPTM4B-induced stemness activation of CD133^+ CSLCs is inhibited by
   WNT inhibition.

   (A,B) Sphere formation assays in LAPTM4B-overexpressing CD133^+ Huh7
   cells treated with or without XAV-939 (WNT pathway inhibitor). (C–G)
   qPCR and western blots of (C,D) β-catenin and (E–G) EpCAM, SOX9, OCT4,
   and NANOG expression in LAPTM4B-overexpressing CD133^+ Huh7 cells
   treated with or without XAV-939. Data are means ± SD for at least two
   independent experiments, analyzed with unpaired Student’s t tests: ∗p
   <0.05; ∗∗p <0.01; ∗∗∗p <0.001; n.s., no significant difference. Scale
   bars: 200 μm. CSLC, cancer stem-like cell; LAPTM4B, lysosome-associated
   protein transmembrane-4β; WNT, wingless/integrated.

Abnormally high expression of LAPTM4B and β-catenin in patients with HCC
confers a poor prognosis

   According to the above results, LAPTM4B overexpression increases
   β-catenin stability by suppressing its degradation. Furthermore, the
   activity of WNT/β-catenin signaling is required for LAPTM4B-induced
   CD133^+ CSLC stemness. Thus, we analyzed their expression patterns in
   surgical samples obtained from patients with HCC to additionally
   investigate the association between LAPTM4B and β-catenin in HCC and
   their potential implications for clinical prognosis.

   The TCGA database was used to identify the mRNA levels of LAPTM4B and
   β-catenin, between which there was a positive correlation (R = 0.18;
   p = 0.0004) ([147]Fig. 7A). IHC analysis revealed that elevated LAPTM4B
   protein levels were associated with the accumulation of β-catenin in
   the nucleus in tissue microarrays (TMAs) comprising 105 HCC specimens
   from a local cohort with long-term clinical follow-up data
   ([148]Fig. 7B,C). A positive correlation between LAPTM4B and β-catenin
   was observed (p <0.05; [149]Fig. 7D; [150]Table S1).

Fig. 7.

   [151]Fig. 7
   [152]Open in a new tab

   LAPTM4B expression is correlated with β-catenin in HCC tissues.

   (A) Correlation between LAPTM4B and β-catenin expression in TCGA
   database confirmed by Pearson’s correlation analysis. (B)
   Representative images of LAPTM4B and β-catenin expression based on
   immunohistochemistry of tissue microarrays. (C) Expression level of
   LAPTM4B and β-catenin in HCC tissues indicated as negative (–), weak
   (+), moderate (++), and strong (+++) expression. (D) Pearson’s
   correlation analysis of LAPTM4B and β-catenin expression in patients
   with HCC. (E,F) Kaplan-Meier curves for overall survival of patients
   with HCC with high or low LAPTM4B/β-catenin expression. (G)
   Kaplan-Meier curves for overall survival of patients with high or low
   levels of LAPTM4B and β-catenin. (H) Working model. Survival rate was
   estimated by the Kaplan-Meier method and groups were compared by Log
   rank test. Scale bars: 100 μm. HCC, hepatocellular carcinoma; LAPTM4B,
   lysosome-associated protein transmembrane-4β; TCGA, the Cancer Genome
   Atlas.

   In addition, we obtained and extracted eight clinical factors from
   clinical follow-up records relating to the clinical background of the
   patients (sex, age, presence of HBV, and cirrhosis) and information
   about cancer stage (tumor size, microvascular invasion, Edmondson
   classification, and microsatellites). We found that increased LAPTM4B
   was only associated with tumor size and Edmondson classification (p
   <0.05, [153]Table S2).

   Subsequently, a Kaplan-Meier analysis was conducted to assess the
   correlation between LAPTM4B and β-catenin expression levels and the
   overall survival of patients. The findings indicated a significant
   relationship between elevated LAPTM4B/β-catenin levels and an
   unfavorable prognosis (p = 0.0026 and p = 0.0388, respectively)
   ([154]Fig. 7E,F). Furthermore, individuals with heightened LAPTM4B and
   β-catenin expression exhibited reduced survival time, whereas those
   with lower levels of these proteins demonstrated prolonged survival
   (p = 0.0043) ([155]Fig. 7G). The 3-year overall survival rate was
   significantly decreased in patients with HCC with high LAPTM4B and
   β-catenin expression compared with those with low expression (42.9% vs.
   74.4%, HR 5.174, 95% CI 2.280–11.741, p <0.001, [156]Table S3). In
   summary, these results emphasize that overactivation of LAPTM4B and
   β-catenin indicates a negative prognosis in patients with HCC. They
   also reveal a significant pathological connection between LAPTM4B and
   β-catenin expression and the development of HCC.

Discussion

   Cancer stem cells significantly contribute to tumor initiation,
   progression, metastasis, recurrence, and drug
   resistance.[157]^3^,[158]^24^,[159]^33 Given the emergence of targeted
   therapies, the concept of cancer stem cells offers novel perspectives
   for diagnosing and predicting the prognosis of tumors. However, the
   molecular mechanisms by which cancer stem cells regulate and maintain
   their stemness characteristics remain unclear, and the clinical
   application of targeted therapy is subject to various limitations.
   Therefore, exploring the regulatory mechanism of stemness is
   particularly important. In this study, we identified that LAPTM4B
   enhances CD133^+ CSLCs stemness through the WNT/β-catenin pathway.

   Overexpression of LAPTM4B results in uncontrolled cell proliferation
   and malignant transformation, which could contribute to HCC progression
   as an oncogene.[160]^34^,[161]^35 Previous studies demonstrated that
   LAPTM4B can enhance the proliferation and invasion of HCC cells, while
   maintaining their stemness through a positive feedback loop with
   Yes-related protein (YAP), thereby promoting HCC
   progression.[162]^8^,[163]^20 Here, we further investigated WNT
   signaling, a regulatory pathway by which LAPTM4B promotes stemness in
   HCC.

   WNT/β-catenin signaling is commonly activated in HCC and is crucial for
   regulating stem cell renewal.[164]^36 In this study, we revealed
   additional evidence supporting the role of LAPTM4B in enhancing the
   functional capacity of the WNT/β-catenin pathway by inhibiting
   β-catenin phosphorylation and ubiquitination degradation, as well as
   its involvement in regulating stemness in HCC. In addition, our study
   provides evidence confirming that the combined high expression of
   LAPTM4B and β-catenin is associated with a poor prognosis for patients
   with HCC. Thus, targeting the LAPTM4B–WNT/β-catenin axis could improve
   the antitumor efficacy of drugs by attenuating tumor cell
   proliferation, invasion, and stemness in HCC. Encouragingly, various
   small-molecule inhibitors of the WNT/β-catenin pathway are now
   commercially available. E7386 is in phase I clinical trials as the
   first-in-class orally active β-catenin-CBP antagonist inhibiting the
   WNT/β-catenin pathway in tumor xenograft models derived from patients
   with HCC.[165]^37 Nonetheless, more comprehensive clinical trials are
   needed to further evaluate potential targets and explore personalized
   treatment options.

   In conclusion, this study revealed that LAPTM4B activates the
   WNT/β-catenin pathway, which is involved in regulating HCC stemness.
   Our study confirms that LAPTM4B inhibits β-catenin phosphorylation.
   Furthermore, LAPTM4B interacts with USP1 and USP14 to inhibit the
   ubiquitinated degradation of β-catenin. This contributes to increased
   β-catenin stability and promotes β-catenin nuclear localization.
   Accumulation of β-catenin in the nucleus promotes activation of the
   WNT/β-catenin pathway and enhances tumor stemness. Abnormal
   upregulation of LAPTM4B and β-catenin enhances stemness and correlates
   with a poorer HCC prognosis ([166]Fig. 7H). The findings of our study
   provide insights into stemness regulation in HCC and potential
   therapeutic targets for individuals with this condition.

Abbreviations

   APC, adenomatous polyposis coli; c-MYC, cellular myelocytomatosis;
   CD133, cluster of differentiation 133; CHX, cycloheximide; Co-IP,
   co-immunoprecipitation; CSLC, cancer stem-like cell; EMT,
   epithelial-to-mesenchymal transition; GEPIA, gene expression profiling
   interactive analysis; GSK-3β, glycogen synthase kinase 3β; HCC,
   hepatocellular carcinoma; HR, hazard ratio; IF, immunouorescent
   staining; IHC, immunohistochemistry; LAPTM4B, lysosome-associated
   protein transmembrane-4β; MMP7, matrix metalloproteinase 7; p-GSK-3β,
   phosphorylated glycogen synthase kinase 3β; qPCR, quantitative
   real-time PCR; RNA-seq, RNA sequencing; STRING, Search Tool for the
   Retrieval of Interacting Genes/Proteins; TCGA, The Cancer Genome Atlas;
   TMAs, tissue microarrays; USP, ubiquitin carboxyl-terminal hydrolase;
   WNT, wingless/integrated; YAP, Yes-related protein.

Financial support

   This project was supported by the Startup Fund for Scientific Research,
   Fujian Medical University (Grant number: 2022QH2004) and the Foundation
   of the Finance Department of Fujian Province of China (Grant/Award
   Number: 21SCZZX002).

Authors’ contributions

   Concept and design: AMH, JPL, and JHW. Performed experiments and
   analyzed data: JHW, JPL, YC, and MRC. Collected human tissues and
   clinical data: JHW and YC. Animal experiments: JHW and JPL. Wrote the
   manuscript: JHW, JPL, and AMH. Reviewed the manuscript: all authors.

Data availability statement

   The authors confirm that the data supporting the findings of this study
   are available within the article and/or its supplementary materials and
   methods. Any additional data are available after approval from the
   corresponding author.

Conflicts of interest

   The authors declare no conflicts of interest.

   Please refer to the accompanying ICMJE disclosure forms for further
   details.

Footnotes

   Author names in bold designate shared co-first authorship

   Supplementary data to this article can be found online at
   [167]https://doi.org/10.1016/j.jhepr.2024.101306.

Supplementary data

   The following are the supplementary data to this article:
   Multimedia component 1
   [168]mmc1.pdf^ (3.2MB, pdf)
   Multimedia component 2
   [169]mmc2.docx^ (55.5KB, docx)
   Multimedia component 3
   [170]mmc3.pdf^ (400.2KB, pdf)
   Multimedia component 4
   [171]mmc4.pdf^ (6.9MB, pdf)

References