TRIM17-mediated ubiquitination and degradation of RBM38 promotes cisplatin resistance in non-small cell lung cancer

Abstract

Cisplatin (CDDP)-based chemotherapy is commonly used to treat advanced non-small cell lung cancer (NSCLC). However, the efficacy is limited by the development of drug resistance. Tripartite motif (TRIM) proteins typically have E3 ubiquitin ligase activities and modulate protein stability. In the present study, we screened for chemosensitivity-regulating TRIM proteins using CDDP-resistant NSCLC cell lines. We show that TRIM17 is upregulated in CDDP-resistant NSCLC cells and tumors compared to CDDP-sensitive counterparts. NSCLC patients with high TRIM17 expression in tumors have shorter progression-free survival than those with low TRIM17 expression after CDDP chemotherapy. Knockdown of TRIM17 increases the sensitivity of NSCLC cells to CDDP both in vitro and in vivo. In contrast, overexpression of TRIM17 promotes CDDP resistance in NSCLC cells. TRIM17-mediated CDDP resistance is associated with attenuation of reactive oxygen species (ROS) production and DNA damage. Mechanistically, TRIM17 interacts with RBM38 and promotes K48-linked ubiquitination and degradation of RBM38. TRIM17-induced CDDP resistance is remarkably reversed by RBM38. Additionally, RBM38 enhances CDDP-induced production of ROS. In conclusion, TRIM17 upregulation drives CDDP resistance in NSCLC largely by promoting RBM38 ubiquitination and degradation. Targeting TRIM17 may represent a promising strategy for improving CDDP-based chemotherapy in NSCLC.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13402-023-00825-6.

Introduction

Lung cancer is the most frequent cause of cancer-related death in the world [1]. Non-small cell lung cancer (NSCLC) is the main histological subtype and constitutes over 80% of all lung cancers. Surgery is a potential curative treatment option for lung cancer. The prognosis for lung cancer patients is still dismal, mostly because of the development of distant metastatic disease [2, 3]. Cytotoxic chemotherapy is widely used to treat advanced NSCLC. An open-label randomized trial demonstrates that patients with advanced non-squamous NSCLC receiving different chemotherapy regimens have a median overall survival of about 1 year [4]. Nevertheless, the survival benefit obtained from chemotherapy is limited, largely due to the emergence of drug resistance [5].

Cisplatin (CDDP) is the backbone of first-line chemotherapy for many types of cancers including NSCLC [6]. The antitumor activity of CDDP is usually linked to induction of DNA damage [7]. CDDP can promote the release of excessive reactive oxygen species (ROS), which cause detrimental effects on tumor cells [8]. Several mechanisms have been suggested to mediate resistance to CDDP, such as overexpression of drug transporters, prevention of DNA binding, enhancement of DNA repair, and activation of cell survival signaling pathways [9]. Yet, the exact mechanism responsible for CDDP resistance in NSCLC is not fully understood.

Tripartite motif (TRIM) proteins comprise a large protein family, most of which have a RING-finger domain, 2 B-box-type zinc fingers, and a coiled-coil domain. To date, more than 80 TRIM family members have been identified in humans. TRIMs commonly possess E3 ubiquitin ligase activities [10]. They can bind with target proteins, leading to ubiquitination and subsequent proteasomal degradation [11]. A few TRIM proteins have been found to be involved in cancer development and progression [12–14]. For instance, TRIM22 has the capacity to enhance glioblastoma growth through induction of IκBα degradation and activation of NF-κB signaling [12]. Depletion of TRIM29 impairs cancer stem cell-like features of pancreatic ductal adenocarcinomas [14]. Therefore, TRIMs are able to coordinate multiple aspects of tumor biology.

It has been documented that TRIM65 and TRIM23 function as a regulator of CDDP sensitivity in NSCLC cells [15, 16]. Knockdown of TRIM65 inhibits autophagy and overcomes CDDP resistance in CDDP-resistant A549 lung cancer cells [15]. Elevated TRIM23 confers survival advantages to A549 cells by regulating glucose metabolism in response to CDDP treatment [16]. To identify novel chemosensitivity-regulating TRIM proteins, we examined a panel of TRIM genes in parental and CDDP-resistant NSCLC cells and revealed the upregulation of TRIM17 in CDDP-resistant NSCLC cells. The clinical significance, biological function, and molecular mechanism of TRIM17 in NSCLC were further interrogated.

Materials and methods

Cell culture and treatment

NSCLC cell lines (A549, H1299, H358, and H460) were purchased from the American Type Culture Collections (ATCC, Rockville, MD, USA). They were cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin–streptomycin, and 2 mM L-glutamine (Sigma-Aldrich, St. Louis, MO, USA). Human embryonic kidney 293 (HEK293) cells were also grown in DMEM with 10% FBS. All cell lines were authenticated using short tandem repeat (STR) profiling and tested negative for mycoplasma contamination. For inhibition of ROS accumulation, cells were treated with N-acetylcysteine (NAC; Sigma-Aldrich) at 2 mM for 2 h.

CDDP-resistant cell lines (A549/CDDP and H460/CDDP) were derived from their parental cells as described previously [17]. In brief, A549 and H460 cells were exposed to increasing concentrations of CDDP (Sigma-Aldrich) from 1 to 15 μM. After 6-month continuous treatment, CDDP-resistant cell lines were generated and maintained in culture media containing 5 μM CDDP.

Cell viability assay

Cells were seeded in 96-well plates and cultured overnight. The cells were then treated with indicated concentrations of CDDP for 72 h. For assessment of viability, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich) was added and incubated for 4 h. After dissolving the resultant formazan crystals in dimethyl sulfoxide, absorbance was measured at 595 nm. The half maximal inhibitory concentration (IC₅₀) for CDDP was calculated.

Apoptosis analysis

Apoptosis was analyzed by an enzyme-linked immunosorbent assay (ELISA) method, which is applied to detect histone-associated DNA fragments. The Cell Death Detection ELISA kit (Roche Applied Science, Indianapolis, IN, USA) was used in this experiment. In brief, cells were seeded in 96-well plates (4 × 10³ cells/well) and treated with indicated concentrations of CDDP for 48 h. The cells were lysed and centrifuged, and the supernatant was added to a strepavidin-coated microtitre plate. After incubation with anti-histone biotin and anti-DNA peroxidase antibodies, each well of the plate was added with the peroxidase substrate. Absorbance was measured at 405 nm.

Colony formation assay

Cells were seeded in 6-well plates at 600 cells/well and treated with indicated concentrations of CDDP for 72 h. Surviving colonies were fixed after 14 days and stained with 0.05% crystal violet. The number of colonies was counted.

Quantitative real-time PCR (qRT-PCR) analysis

Total RNA was extracted with TRIzol reagent (Thermo Fisher Scientific, Carlsbad, CA, USA). Reverse transcription was completed using the High Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). qRT-PCR was performed using the iTaq Universal SYBR Green One-Step Kit (Bio-Rad, Hercules, CA, USA) as per the manufacturer's instructions. Primers used for qRT-PCR are summarized as follows: TRIM17 sense: 5′-GACATGGAGTACCTTCGGGA-3′, TRIM17 antisense: 5′-GCAGTCTCCTCTTCTTCCGT-3′; RBM38 sense: 5′-AAGACCCGAACCCCATCATC-3′, RBM38 antisense: 5′-CACGATGGCTGGTGGGTAGA-3′; GAPDH sense: 5′-TCTCCTCTGACTTCAACAGC-3′, GAPDH antisense: 5′-CTGTTGCTGTAGCCAAATTCG-3′; ACTB sense: 5′-GATGCGTTGTTACAGGAAGTCC-3′, ACTB antisense: 5′-GGCACGAAGGCTCATCATTCA-3′; TBP sense: 5′-GAGCTGTGATGTGAAGTTTCC-3′, TBP antisense: 5′-TCTGGGTTTGATCATTCTGTAG-3′. Primer sequences for the other TRIM family genes tested are available on request. The levels of the mRNAs of interest were normalized to the average value of the 3 housekeeping genes (i.e. GAPDH, ACTB and TBP) and calculated using the the 2^−ΔΔCt method.

Patients and tissue samples

We collected 80 paraffin-embedded bronchoscopic/fine needle aspiration biopsy samples from patients with advanced NSCLC (stage III or IV) who received CDDP-based chemotherapy as the first-line therapy between 2010 and 2013. The characteristics of the 80 patients enrolled are summarized in Supplementary Table S1. None of the tumor samples showed activating epidermal growth factor receptor (EGFR) mutations. All the patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. They were given 2–6 cycles of 75 mg/m² CDDP and 500 mg/m² pemetrexed on day 1 every 3 weeks as induction therapy. For patients who had a complete response (CR), partial response (PR), or experienced stable disease (SD) from induction chemotherapy, 500 mg/m² pemetrexed was administered as maintenance therapy on day 1 of 21-day cycles until progressive disease (PD) or treatment discontinuation. The median follow-up period was 6 months (range 2–24 months). Tumor response was assessed as per the Response Evaluation Criteria in Solid Tumors (version 1.1) by computed tomography every 6 weeks after the onset of the induction chemotherapy until PD. Chemotherapeutic resistance to CDDP was defined as response less than PR, SD or PD after induction therapy with CDDP plus pemetrexed. The patients showing CR or PR to the CDDP-based induction chemotherapy were considered ‘sensitive’. Progression-free survival was calculated from the start date of chemotherapy to disease progression or the time of the last follow-up. Tumor samples were fixed in formalin, embedded in paraffin, and sectioned. The tissue sections were then processed for immunohistochemical analysis of TRIM17.

Immunohistochemical analysis

Paraffin-embedded tissue sections were deparaffinized, rehydrated, and exposed to 3% H₂O₂ to eliminate endogenous peroxidase. Pretreatment was done in microwave with Tris/EDTA buffer solution (pH 8.0). After blocking with 5% normal goat serum, tissue sections were incubated with anti-TRIM17 antibody (ab235527, Abcam, Cambridge, MA, USA; 1:100 dilution). Immunoreaction was detected using the EnVision Dako system (Dako, Milan, Italy). The sections were counterstained with hematoxylin. Human colorectal cancer tissue samples were used as a positive control. Phosphate buffered saline was used as a negative control by substituting for anti-TRIM17 antibody.

TRIM17 immunoreactivity was evaluated by one experienced pathologist in a blind manner. The intensity of immunostaining was graded on the following scale: 0, no staining; 1, weak; 2, moderate; 3, strong. The area of immunostaining was scored as follows: 0, < 10% of cells stained positive; 1, 10–25%; 2, 26–50%; and 3, > 50%. The histochemical score was yielded by multiplying immunostaining intensity by immunostaining percentage [18]. We used 4 as the cut-off score to distinguish high and low TRIM17 expression tumors.

Plasmids, siRNAs, and transfections

The plasmids expressing TRIM17 (GenBank no. NM_016102) and RBM38 (GenBank no. NM_017495) were generated by inserting corresponding full-length cDNA sequences into the pcDNA3.1( +) vector. A truncated TRIM17 fragment with deletion of the RING domain (TRIM17-ΔRING) was inserted into the pcDNA3.1/N-Myc vector. His-tagged ubiquitin and ubiquitin lysine mutants were described as previously [19]. Flag-tagged RBM38 was constructed by cloning RBM38 cDNA into the pCMV3-N-FLAG vector. Short hairpin RNAs (shRNAs) targeting 2 different sites of TRIM17 were cloned into the pLKO.1 vector, with the target sequences listed as follows: shTRIM17#1, 5′-GAGCCTGCATCCAGCTGAGCT-3′, and shTRIM17#2, 5′-GTTCTGCAGCAAGGACCGATT-3′. The sequence of a negative control shRNA (shCtrl) was as follows: 5′-CAACAAGATGAAGAGCACCAA-3′. A pool of 3 small interfering RNAs (siRNAs) were used to target RBM38, with the sense sequences shown below: siRBM38#1, 5′-UGAGAGGGCUUGCAAAGAC-3′, siRBM38#2, 5′-GACACCACGUUCACCAAGA-3′, and siRBM38#3, 5′-ACGCCUCGCUCAGGAAGUA-3′.

Transfection of indicated plasmids into NSCLC cells was achieved using Lipofectamine 3000 transfection reagent as per the manufacturer's instructions (Thermo Fisher Scientific). For generation of stable cell lines, NSCLC cells transfected with shTRIM17#2 or TRIM17-expressing plasmid were selected with puromycin (2 μg/mL; Thermo Fisher Scientific) or G418 (800 μg/mL; Sigma-Aldrich), respectively.

Animal experiments

Five-week-old male BALB/c nude mice were acclimated for 1 week with free access to food and water. The mice were randomly assigned and subcutaneously injected with 1 × 10⁷ tumor cells stably expressing TRIM17 or shTRIM17#2 (n = 4). When xenograft tumors reached a volume of ~ 100 mm³, CDDP (2 mg/kg body weight) was intraperitoneally given for 5 consecutive days. Tumors were measured every 5 days until 25 days after cell injection. Tumor volume was calculated using the following formula: tumor volume (mm³) = length (mm) × width² (mm²) × 0.5. After the last measurement, all mice were euthanized. Data were collected in a blind manner.

Measurement of intracellular ROS levels

Intracellular ROS levels were measured using a cell permeant reagent 2’-7’dichlorofluorescin diacetate (DCFH-DA), which can be oxidized by intracellular ROS to yield a highly fluorescent product 2’-7’dichlorofluorescein (DCF). In brief, NSCLC cells (2 × 10⁴ cells/well) were seeded in 96-well plates and incubated with 10 μM of DCFH-DA (Cell Biolabs, San Diego, CA, USA) for 30 min at 37 °C in the dark. After removal of DCFH-DA, cell lysis buffer was added to each well and incubated for 5 min. The lysate was tested for fluorescence intensities at 530 nm using a fluorometric plate reader.

Immunofluorescent staining

γH2AX foci formation was analyzed to evaluate DNA damage, as described previously [20]. Briefly, cells were seeded on glass coverslips in 24-well plates (1 × 10⁴ cells/well) and treated with indicated concentrations of CDDP for 48 h. The cells were then fixed with 4% paraformaldehyde and incubated with anti-phospho-γH2AX (ab26350, Abcam; 1:50 dilution), followed by incubation with Alexa Fluor 568-conjugated goat anti-rabbit IgG. Nuclei were counterstained with Hoechst 33342 (Thermo Fisher Scientific). The percentage of γH2AX positive cells was determined. For TRIM17 and RBM38 double staining, cells were incubated with rabbit anti-TRIM17 antibody (ab235527, Abcam) and mouse anti-RBM38 antibody (ab168445, Abcam). Alexa Fluor 568-conjugated goat anti-rabbit and fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse secondary antibodies were then used. After nuclei staining, the cells were imaged by a confocal fluorescence microscope.

Immunoprecipitation and Western blot analysis

Cells were lysed in ice-cold radioimmunoprecipitation assay buffer (RIPA) containing 25 mM Tris–HCl (pH 7.6), 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate, and 0.1% sodium dodecyl sulfate (SDS), which was supplemented with a protease inhibitor cocktail (Cell Signaling Technology, Beverly, MA, USA). Protein concentration was measured using a BCA protein assay kit (Beyotime, Beijing, China). For immunoprecipitation assays, lysates were pre-cleared with rabbit IgG for 2 h, followed by incubation overnight at 4 °C with 6 μg anti-TRIM17 (ab235527, Abcam), anti-RBM38 (ab168445, Abcam), anti-HA (ab1424, Abcam), or anti-Flag (#14793, Cell Signaling Technology) antibodies. The immune complexes were captured by Protein A/G agarose beads (Beyotime). After washing, immunoprecipitated proteins were eluted with 0.2 M glycine (pH 2.0). Eluted proteins were digested with trypsin overnight at 37 °C. The resulting tryptic peptides were analyzed by tandem mass spectrometry. Peptide sequence analysis was performed by searching against a UniProt protein database.

For Western blot analysis, the following primary antibodies were used, i.e., anti-TRIM17 antibody (ab235527, Abcam), anti-RBM38 antibody (ab168445, Abcam), anti-ubiquitin (ab140601, Abcam), anti-HA (ab1424, Abcam), anti-Flag (#14793, Cell Signaling Technology), anti-Myc-Tag (#2276, Cell Signaling Technology), and anti-GAPDH antibody (ab181602, Abcam). After incubation with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling Technology), protein signals were visualized with the ECL Plus Chemiluminescence Detection Kit (Thermo Fisher Scientific).

Analysis of RBM38 protein turnover

To analyze RBM38 protein turnover, NSCLC cells were treated with 40 μg/mL cycloheximide (CHX; Sigma-Aldrich) and then harvested at different time points. The cells were lysed and measured for RBM38 protein levels by Western blot analysis.

Ubiquitination assay

Ubiquitination assays were performed under denaturing conditions, as described previously [21]. In brief, cells were transfected with His-ubiquitin (wild-type or different mutants), Flag-RBM38, and Myc-TRIM17 (wild-type or an inactive mutant). The cells were treated with 20 μM MG132 (Sigma-Aldrich) for 8 h and lysed in guanidium-HCl lysis buffer (6 M guanidine-HCl, pH 8.0, 0.1 M Na₂HPO₄/NaH₂PO₄, and 10 mM imidazole). The lysates were incubated with nickel-nitrilotriacetic acid (Ni–NTA) agarose for 3 h at 4 °C. The pull-down products were washed and subjected to Western blot analysis.

Statistical analysis

All in vitro experiments were repeated three times unless stated otherwise. Each assay had a proper sample size. The comparison of statistical differences was achieved using the Student’s t test or one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Normality of distribution and homogeneity of variance were examined. Survival curves was estimated using the Kaplan–Meier method and compared using log-rank test. A multivariate Cox proportional hazards regression model was generated to evaluate the prognostic significance of TRIM17 expression for progression-free survival of advanced NSCLC patients. P-values < 0.05 were considered significant.

Results

Elevated expression of TRIM17 correlates with CDDP resistance in NSCLC cells and patients

To search for TRIM members involved in CDDP resistance, we compared the transcript levels of 82 TRIM genes in parental and CDDP-resistant H460 and A549 cells using qRT-PCR analysis. As illustrated in Fig. 1A and Supplementary Fig. S1, 4 TRIMs showed significant differential expression in H460/CDDP cells and 9 in A549/CDDP cells. Most interestingly, TRIM17 was significantly induced in both the CDDP-resistant cell lines compared to their parental counterparts, i.e., with 5.85- and 3.92-fold increase in H460/CDDP and A549/CDDP cells, respectively (Fig. 1B). TRIM17 protein levels were also increased in CDDP-resistant cells relative to parental cells (Fig. 1C). Next, we evaluated the relationship between TRIM17 expression and CDDP resistance in NSCLC patients. We found that TRIM17 staining score was remarkably higher in tumor samples from CDDP refractory patients than those from CDDP sensitive patients (Fig. 1D and E). With regard to treatment outcomes, high expression of TRIM17 in tumors was associated with reduced progression-free survival after CDDP chemotherapy (P < 0.0001; Fig. 1F). Multivariate Cox regression analysis demonstrated that TRIM17 expression (high vs. low, hazard ratio (HR) = 1.42, 95% confidence interval (CI) = 1.07–3.55, P = 0.0192) and TNM stage (IV vs. III, HR = 2.37, 95% CI = 1.49–4.96, P = 0.0045) were significant prognostic factors for progression-free survival of advanced NSCLC patients (Table 1). These results have suggested an association between TRIM17 upregulation and CDDP resistance of NSCLC.

Fig. 1.

Elevated expression of TRIM17 correlates with CDDP resistance in NSCLC cells and patients. (A) Venn diagram showing the number of genes with significant differential expression between CDDP-resistant and their parental cells. Upward arrows indicate upregulated genes and downward arrows indicate downregulated genes in CDDP-resistant cells. (B) qRT-PCR analysis of TRIM17 mRNA levels in H460/CDDP, A549/CDDP, and their parental cells. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (C) Western blot analysis of TRIM17 protein levels in H460/CDDP, A549/CDDP, and their parental cells. (D) Immunohistochemical analysis of TRIM17 expression in tumor samples from CDDP-resistant (n = 49) and CDDP-sensitive (n = 31) NSCLC patients. Scale bar = 100 μm. (E) Immunohistochemical scores of TRIM17. ^*P < 0.05. (F) Kaplan–Meier curves were estimated to evaluate the association between TRIM17 expression and progression-free survival for NSCLC patient receiving CDDP chemotherapy

Table 1.

Multivariate Cox regression analysis of prognostic factors associated with progression-free survival

Characteristic	Hazard ratio	95% CI	P
Gender (male vs. female)	0.94	0.51–1.57	0.6128
TNM stage (IV vs. III)	2.37	1.49–4.96	0.0045
ECOG PS (0 vs. 1)	0.79	0.36–1.22	0.2734
TRIM17 expression (high vs. low)	1.42	1.07–3.55	0.0192

CI confidence interval; ECOG PS Eastern Cooperative Oncology Group performance status

TRIM17 is required for maintenance of CDDP resistance

To clarify the biological role of TRIM17 in CDDP resistance, we stably knocked down TRIM17 expression in CDDP-resistant cells by transfecting with 2 different shRNAs (Fig. 2A and B). Notably, targeted reduction of TRIM17 led to a profound suppression of cell viability upon exposure to 15 μM CDDP (Fig. 2C). However, silencing of TRIM17 did not directly affect cell viability. Analysis of DNA fragmentation, a hallmark of apoptosis revealed that depletion of TRIM17 significantly enhanced CDDP (15 μM)-induced apoptosis in A549/CDDP and H460/CDDP cells (Fig. 2D). TRIM17-depleted cells exhibited a remarkable reduction in IC₅₀ values for CDDP, compared to control cells (Fig. 2E). Moreover, colony formation capabilities were attenuated in TRIM17-depleted cells after CDDP (15 μM) treatment (Fig. 2F and G). We also evaluated the effect of TRIM17 depletion on parental NSCLC cells. Similar to the findings seen in CDDP-resistant cells, depletion of TRIM17 significantly increased the sensitivity of A549 and H460 cells to CDDP (Supplementary Fig. S2).

Fig. 2.

Depletion of TRIM17 reverses CDDP resistance in CDDP-resistant cells. (A) Knockdown of TRIM17 in A549/CDDP and H460/CDDP cells through transfection of 2 different shRNAs (shTRIM17#1 and #2). Data are represented as mean ± SD (n = 3). ^*P < 0.05. (B) Western blot analysis of TRIM17 protein levels in A549/CDDP and H460/CDDP cells transfected with indicated constructs. (C) Measurement of viability in NSCLC cells transfected with indicated constructs and treated with or without 15 μM CDDP for 72 h. Results are expressed as percentage of viability of shCtrl-transfected cells without CDDP treatment. ^*P < 0.05 (n = 3). ns indicates no significance. (D) Analysis of DNA fragmentation, a hallmark of apoptosis by ELISA. The results are expressed as fold change in apoptosis relative to shCtrl-transfected cells without CDDP treatment. Data are represented as mean ± SD (n = 3). ^*P < 0.05. ns indicates no significance. (E) A549/CDDP and H460/CDDP cells transfected with indicated constructs were treated with increasing concentrations of CDDP for 72 h, and the IC₅₀ values for CDDP were determined. (F,G) Colony formation assay comparing A549/CDDP and H460/CDDP cells transfected with indicated constructs and treated with or without 15 μM CDDP. Graphs in (F) show the formation of colonies after culturing for 14 days. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (H,I) Depletion of TRIM17 enhances CDDP sensitivity in vivo. Macroscopic images in (H) show one representative xenograft tumor from each group. (I) Comparison of tumor growth curves for each group (n = 4). ^*P < 0.05

TRIM17 knockdown-mediated CDDP sensitization was then validated in vivo. We subcutaneously injected CDDP-resistant A549 cells with or without TRIM17 depletion into nude mice. It was found that xenograft tumors generated from TRIM17-depleted cells were significantly more sensitive to CDDP than control tumors (Fig. 2H and I). These results indicate that TRIM17 plays an essential role in the maintenance of CDDP resistance.

TRIM17 overexpression confers CDDP resistance to NSCLC cells

Next, we asked whether TRIM17 overexpression could facilitate the development of CDDP resistance in NSCLC. To this end, we generated stable A549 and H460 cells overexpressing TRIM17 and treated them with CDDP. Overexpression of TRIM17 attenuated CDDP-induced cytotoxicity in a series of NSCLC cell lines including A549, H1299, H358, and H460 cells (Fig. 3A-C and Supplementary Fig. S3). The IC₅₀ values for CDDP were significantly elevated in TRIM17-overexpressing cells relative to control cells. After long-term CDDP (10 μM) treatment, TRIM17-overexpressing cells formed significantly more colonies than control cells (Fig. 3D and E). In vivo xenograft tumor studies confirmed that TRIM17 overexpression rendered A549 xenograft tumors more resistant to CDDP treatment (Fig. 3F and G). These findings demonstrate that TRIM17 functions as an inducer of CDDP resistance.

Fig. 3.

TRIM17 overexpression induces CDDP resistance in NSCLC cells. (A) Western blot analysis of TRIM17 protein levels in A549 and H460 cells transfected with indicated constructs. (B) TRIM17-overexpressing cells had higher IC₅₀ values for CDDP than control cells. Data are shown as mean ± SD of three independent experiments. (C) Analysis of apoptosis by ELISA. The results are expressed as fold change in apoptosis relative to vector-transfected cells without CDDP treatment. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (D, E) Colony formation assay. A549 and H460 cells transfected with indicated constructs were treated with or without 10 μM CDDP for 72 h. The formation of colonies was determined after culturing for additional 14 days. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (F, G) Overexpression of TRIM17 promotes CDDP resistance in a mouse model. Macroscopic images in (F) show one representative xenograft tumor from each group. Tumor growth curves for each group were plotted based on tumor volume and compared. ^*P < 0.05 (n = 4)

TRIM17 protects from CDDP-induced DNA damage and ROS overproduction

CDDP has been known to cause cytotoxic effects by inducing DNA damage and ROS overproduction [8, 22]. Hence, we tested whether TRIM17 has an impact on DNA damage repair and ROS formation. CDDP (10 μM) treatment led to formation of γH2AX foci in a substantial number of A549 and H460 cells (Fig. 4A and B). When TRIM17 was overexpressed, the number of CDDP-induced γH2AX foci was significantly lowered. Moreover, CDDP-induced ROS production was remarkably counteracted by TRIM17 overexpression (Fig. 4E). In contrast, knockdown of TRIM17 aggravated DNA damage and augmented ROS generation in CDDP-resistant cells in the presence of 15 μM CDDP (Fig. 4C, D and F). Taken together, TRIM17 contributes to the development of CDDP resistance in NSCLC.

Fig. 4.

TRIM17 protects from CDDP-induced DNA damage and ROS overproduction. (A,B) γH2AX immunostaining performed in (A) A549 and (B) H460 cells transfected with TRIM17-expressing plasmid and treated with or without 10 μM CDDP for 48 h. Scale bar = 25 μm. (C, D) γH2AX immunostaining performed in CDDP-resistant cells transfected with TRIM17-targeting shRNA (shTRIM17#2) and treated with or without 15 μM CDDP for 48 h. (E) Measurement of ROS levels in A549 and H460 cells treated as in (A). (F) Analysis of ROS levels in CDDP-resistant cells treated as in (C, D). Data are represented as mean ± SD (n = 3). ^*P < 0.05

TRIM17 promotes CDDP resistance via degradation of RBM38

To ascertain the mechanism involved in TRIM17-induced CDDP resistance, we performed TRIM17 co-immunoprecipitation and characterized the immunoprecipitated proteins by mass spectrometry. This experiment identified RBM38 as a TRIM17-associated protein in both A549 and H460 cells (Supplementary Table S2). Western blot analysis of TRIM17 immunoprecipitates confirmed the association between endogenous TRIM17 and RBM38 (Fig. 5A). Confocal fluorescence microscopy revealed colocalization of TRIM17 and RBM38 in the cytoplasm of A549 cells (Fig. 5B). It has been reported that RBM38 can exert tumor-suppressive effects on multiple cancers such as hepatocellular carcinoma (HCC) and NSCLC [23–25]. We thus examined the role of RBM38 in the regulation of CDDP sensitivity. Intriguingly, RBM38 protein expression was downregulated in CDDP-resistant NSCLC cells, compared to parental cells (Fig. 5C). However, the level of RBM38 transcript was comparable between CDDP-resistant and parental NSCLC cells (Supplementary Fig. S4A). Ectopic expression of RBM38 restored the sensitivity of CDDP-resistant cells to 15 μM CDDP (Fig. 5D-G). Consistently, in vivo studies showed that restoration of RBM38 potentiated CDDP cytotoxicity against xenograft tumors derived from A549/CDDP cells (Fig. 5H). Next, we assessed the clinical significance of RBM38 in NSCLC. Immunohistochemical analysis revealed a significant reduction of RBM38 expression in tumors from CDDP refractory patients, as compared to those from CDDP sensitive patients (Fig. 5I). Moreover, patients in the high RBM38 expression group had significantly longer progression-free survival after CDDP chemotherapy than the low RBM38 group (P = 0.0003; Fig. 5J).

Fig. 5.

RBM38 restores CDDP sensitivity in CDDP-resistant cells. (A) Co-immunoprecipitation assays revealed the association between endogenous TRIM17 and RBM38 in both A549 and H460 cells. (B) Confocal fluorescence microscopy confirmed colocalization of TRIM17 and RBM38 in the cytoplasm of A549 cells. Scale bar = 20 μm. (C) Analysis of RBM38 protein levels in CDDP-resistant cells and their parental controls. (D) Western blot analysis of RBM38 in CDDP-resistant cells after transfection with RBM38-expressing plasmid. (E) Measurement of IC₅₀ values for CDDP in CDDP-resistant cells transfected with indicated constructs. (F, G) Colony formation assay. CDDP-resistant cells were transfected with RBM38-expressing plasmid or vector and treated with or without 15 μM CDDP. Graphs show the formation of colonies after culturing for 14 days. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (H) Overexpression of RBM38 increases CDDP sensitivity in a mouse model. Tumor growth curves for each group were plotted based on tumor volume and compared. ^*P < 0.05 (n = 4). (I) Immunohistochemical analysis of RBM38 expression in tumor samples from CDDP-resistant and CDDP-sensitive NSCLC patients. Scale bar = 100 μm. Right panel showing immunohistochemical scores of RBM38. ^*P < 0.05. (J) Kaplan–Meier curves were estimated to evaluate the association between RBM38 expression and progression-free survival for NSCLC patient receiving CDDP chemotherapy

Having uncovered the chemosensitizing activity of RBM38, we checked whether TRIM17-mediated CDDP resistance is associated with dysregulation of RBM38. Overexpression of TRIM17 led to a reduction of RBM38 protein in both A549 and H460 cells, which was counteracted by the proteasome inhibitor MG132 (Fig. 6A). However, the RBM38 mRNA level remained unchanged in TRIM17-overexpressing cells (Supplementary Fig. S4B). CHX chase assay confirmed that RBM38 protein degradation was accelerated in TRIM17-overexpressing cells (Fig. 6B). These results suggest that TRIM17 overexpression may enhance proteasomal degradation of RBM38. Since TRIM17 has E3 ubiquitin ligase activity [26], we examined the effect of TRIM17 on RBM38 protein ubiquitination status. Ni–NTA pull-down assays performed under denaturing conditions revealed that overexpression of TRIM17 augmented endogenous RBM38 ubiquitination in A549 and H460 cells (Fig. 6C). We also validated the ubiquitination of ectopically expressed RBM38 by TRIM17. As shown in Fig. 6D, overexpression of wild-type TRIM17 promoted RBM38 polyubiquitination in HEK293 cells. However, the inactive mutant TRIM17-ΔRING, which lacked the RING domain required for E3 ligase activity, was unable to induce RBM38 ubiquitination. Deletion of the RING domain completely impaired the ability of TRIM17 to induce CDDP resistance (Supplementary Fig. S5). These results suggest an essential role for the RING domain in mediating the biological activity of TRIM17. In addition, we noted that TRIM17 preferentially catalyzed the formation of K48-linked polyubiquitin chain on RBM38 (Fig. 6E). Rescue experiments further demonstrated that enforced expression of RBM38 restored CDDP sensitivity in TRIM17-overexpressing NSCLC cells (Fig. 6F and G). These results collectively suggest that promotion of RBM38 degradation accounts for TRIM17-induced CDDP resistance.

Fig. 6.

Overexpression of TRIM17 promotes RBM38 polyubiquitination and degradation. (A) Western blot analysis of RBM38 and TRIM17 protein expression in A549 and H460 cells overexpressing TRIM17 with or without treatment with 20 μM MG132. (B) Assessment of RBM38 protein stability in TRIM17-overexpressing NSCLC cells. Cycloheximide (CHX) was used to block protein synthesis, and RBM38 protein levels were quantified at indicated time points. (C) Ubiquitination of RBM38 by TRIM17. NSCLC cells were transfected with indicated plasmids, and the lysates were pulled down by nickel-nitrilotriacetic acid (Ni–NTA) agarose under denaturing conditions and subjected to immnublot (IB) analysis. (D) HEK293 cells were co-transfected with Flag-tagged RBM38 and His-ubiquitin, together with Myc-tagged, wild type (wt) or mutant TRIM17. The ubiquitinated proteins were pulled down by Ni–NTA agarose under denaturing conditions and subjected to IB analysis. (E) Analysis of ubiquitinated proteins pulled down by Ni–NTA agarose in HEK293 cells transfected with Flag-tagged RBM38, Myc-tagged TRIM17, and His-ubiquitin mutants. (F) Measurement of CDDP IC₅₀ values in A549 and H460 cells transfected with indicated constructs. (G) Colony formation assay. A549 and H460 cells were transfected with indicated constructs and treated with or without 10 μM CDDP. Graphs show the formation of colonies after culturing for 14 days. Data are represented as mean ± SD (n = 3). ^*P < 0.05

RBM38 enhances CDDP-induced DNA damage by stimulating ROS generation

Next, we assessed the capacity of RBM38 to regulate DNA damage and ROS production. When RBM38 was knocked down, CDDP (10 μM)-induced DNA damage in A549 and H460 cells was impaired (Fig. 7A-C), suggesting that RBM38 was essential in mediating CDDP cytotoxicity. Overexpression of RBM38 was associated with increased formation of γH2AX foci in CDDP-resistant cells after CDDP (15 μM) treatment (Fig. 7D). Moreover, ROS production was enhanced in RBM38-overexpressing cells after CDDP (15 μM) treatment (Fig. 7E). To check whether excessive ROS generation is involved in RBM38 effects on CDDP-treated lung cancer cells, the ROS scavenger NAC was used to inhibit ROS production. Of note, NAC pretreatment impaired RBM38-mediated enhancement of DNA damage in A549/CDDP and H460/CDDP cells upon CDDP exposure (Fig. 7D), which was accompanied by reduction of ROS accumulation (Fig. 7E).

Fig. 7.

RBM38 enhances CDDP-induced DNA damage by stimulating ROS generation. (A) Measurement of RBM38 mRNA levels in A549 and H460 cells transfected control siRNA (siCtrl) or a pool of 3 different RBM38-targeting siRNAs (siRBM38#1–3). Data are represented as mean ± SD (n = 3). ^*P < 0.05. (B, C) γH2AX immunostaining performed in NSCLC cells transfected with siCtrl or siRBM38#1–3 and treated with or without 10 μM CDDP for 48 h. Scale bar = 25 μm. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (D) γH2AX immunostaining performed in CDDP-resistant cells transiently transfected with vector or RBM38-expressing plasmid. The cells were pretreated with NAC (2 mM) for 2 h before exposure to 15 μM CDDP for 48 h. Data are represented as mean ± SD (n = 3). ^*P < 0.05. (E) Measurement of ROS levels in CDDP-resistant cells treated as in (D). Data are represented as mean ± SD (n = 3). ^*P < 0.05. (F) Schematic model of the TRIM17-RBM38 pathway in regulation of CDDP resistance of NSCLC

Discussion

In this study, we identify TRIM17 as a novel driver of CDDP resistance in NSCLC. Compared to parental control cells, CDDP-resistant NSCLC cells show an induction of TRIM17 expression. Consistently, increased expression of TRIM17 is seen in CDDP-resistant tumor tissues resected from NSCLC patients. Moreover, NSCLC patients with high TRIM17 levels in tumors have significantly shorter progression-free survival after CDDP chemotherapy than those with low TRIM17 levels. These data suggest that TRIM17 may be a potential biomarker for CDDP sensitivity in NSCLC patients. Although several TRIM family members such as TRIM3 [27] and TRIM11 [28] have been found to regulate tumor progression, the role of TRIM17 in malignant disease is not clear. Our results suggest a link between TRIM17 dysregulation and CDDP sensitivity of cancer cells.

The emergence of CDDP resistance is an important cause limiting the long-term survival of cancer patients receiving CDDP chemotherapy [5, 6]. Multiple TRIM proteins have been found to induce CDDP resistance in different cancer types [28–31]. For instance, TRIM11 promotes chemoresistance in nasopharyngeal carcinoma through degradation of Daple [28]. TRIM29 upregulation induces the stem cell-like characteristics and CDDP resistance in ovarian cancer cells [31]. TRIM37 overexpression confers resistance to CDDP in esophageal cancer cells through monoubiquitination of NEMO and activation of NF-κB signaling [29]. In this study, we demonstrate that TRIM17 serve as a novel TRIM family member that can promote CDDP resistance in cancer cells. Enforced expression of TRIM17 renders NSCLC cells more resistant to CDDP cytotoxicity, which explains the association between TRIM17 upregulation and CDDP resistance in NSCLC patients. To evaluate the potential of TRIM17 as a therapeutic target, we knocked down TRIM17 expression in CDDP-resistant NSCLC cells. Of note, knockdown of TRIM17 restores the sensitivity of CDDP-resistant NSCLC cells to CDDP. In vivo studies validate that TRIM17-depleted xenograft tumors show higher sensitivity to CDDP treatment than control tumors. The pro-survival activity of TRIM17 is also observed in melanoma cells [32]. CDDP is a potent inducer of ROS formation and can effectively evoke DNA damage [8, 22]. Attenuation of ROS production upon CDDP treatment has been suggested to contribute CDDP resistance [33, 34]. In our study, TRIM17 shows the ability to block CDDP-induced ROS generation in NSCLC cells. Downregulation of TRIM17 reinforces ROS production provoked by CDDP in CDDP-resistant NSCLC cells. Collectively, TRIM17-mediated CDDP resistance is associated with the promotion of ROS generation.

Lionnard et al. [32] reported that TRIM17 inhibits TRIM28-mediated ubiquitination of an anti-apoptotic factor BCL2A1, thus promoting cell survival. However, the stability of BCL2A1 is not alerted by TRIM17 in NSCLC cells (data not shown), suggesting that TRIM17 may promote CDDP resistance through a BCL2A1-independent mechanism. Previous studies have reported that TRIM17 can regulate the ubiquitination of ZWINT in MCF-7 breast cancer cells [35] and Mcl-1 in neurons [26]. In this article, we reveal that TRIM17 interacts with RBM38 and promotes RBM38 ubiquitination, resulting in enhancement of RBM38 degradation. RBM38 is an important tumor suppressor [36, 37]. Zhang et al. [36] reported that RBM38 deficiency accelerates tumorigenesis in mice. Similarly, ablation of RBM38 promotes lymphomagenesis [38]. In agreement with these studies, we find that RBM38 also has a negative impact on NSCLC aggressiveness. Overexpression of RBM38 increases the sensitivity of CDDP-resistant cells to CDDP. Moreover, RBM38 overexpression reverses CDDP resistance induced by TRIM17. When ROS production is inhibited, RBM38-mediated enhancement of CDDP-elicited DNA damage is blunted. Therefore, induction of ROS production is critical for RBM38-induced CDDP sensitization. Taken together, TRIM17 induces CDDP resistance in NSCLC cells, at least in part through ubiquitination and degradation of RBM38 (Fig. 7F). A previous study has indicated that TRIM17 is able to induce neuronal apoptosis [26]. RBM38 is reported to inhibit apoptosis and enhance adriamycin resistance in breast cancer cells [39]. The role of the TRM17/RBM38 signaling pathway deserves further investigation in modulating cell death.

Several limitations of this study should be noted. Firstly, it remains unclear how TRIM17 is upregulated in CDDP-resistant NSCLC cells. Secondly, further work is needed to clarify the mechanism by which TRM17/RBM38 signaling regulates ROS production and redox homeostasis in NSCLC cells. Finally, while we strengthen the importance of RBM38 in mediating TRIM17 activity, we can not exclude the possibility that other proteins might also contribute to the development of CDDP resistance afforded by TRIM17.

In conclusion, we propose TRIM17 upregulation as a novel mechanism for the development of CDDP resistance in NSCLC cells. The promotion of RBM38 degradation is involved in TRIM17-mediated CDDP resistance. TRIM17 may represent a promising target for overcoming resistance to CDDP-based chemotherapy in NSCLC.

Supplementary Information

Below is the link to the electronic supplementary material.

Authors' contributions

This study was conceived, designed, and interpreted by TZ, JZ and HML. TZ, JZ and XGL undertook the experiments. TZ drafted the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by the Research Foundation from Sichuan Provincial Health and Family Planning Commission of China (17PJ039).

Data availability

All data that support the findings of this study are available from the corresponding author (lihongmin@med.uestc.edu.cn) upon reasonable request.

Declarations

Ethical approval

This study was approved by the Institutional Review Board of University of Electronic Science and Technology of China (Chengdu, China). Written informed consent for research was obtained from each patient. The protocols involving animals were approved by the Institutional Animal Care and User Committee of University of Electronic Science and Technology of China.

Competing interests

The authors declare that they have no conflict of interest.