Long noncoding RNA SNHG5 promotes 5-fluorouracil resistance in colorectal cancer by regulating miR-26b/p-glycoprotein axis

Abstract

BACKGROUND

Colorectal cancer (CRC) is the second most prevalent cause of cancer-related mortality and is increasing in younger individuals. Chemotherapy, a crucial adjuvant systemic therapy for CRC management, often leads to resistance through poorly characterized underlying molecular mechanisms. The long noncoding RNA SNHG5 is highly expressed in CRC and promotes tumor proliferation and invasion, prompting us to hypothesize that SNHG5 may play a crucial role in the chemotherapeutic agent 5-fluorouracil (5-Fu) resistance in CRC.

AIM

To identify the function and mechanism of SNHG5 in 5-Fu resistance in CRC.

METHODS

Quantitative real-time polymerase chain reaction was performed to examine the expression of SNHG5 in CRC tissues from 22 5-Fu-sensitive patients and 14 5-Fu-resistant patients and in CRC cells and 5-Fu-resistant CRC cells. Cell viability and apoptosis were assessed in SNHG5-overexpressing CRC cells and SNHG5-knockdown 5-Fu-resistant CRC cells. SNHG5 function in 5-Fu resistance in CRC was further analyzed using a xenograft mouse model. SNHG5 interactions with microRNAs were predicted by bioinformatics analysis. Luciferase reporter and RNA immunoprecipitation assays were performed to verify the binding between SNHG5 and miR-26b. Rescue experiments were performed to validate the functional interaction between SNHG5 and the miR-26b/p-glycoprotein (Pgp) axis.

RESULTS

SNHG5 expression was upregulated in 5-Fu-resistant CRC tissues and 5-Fu-resistant CRC cells. In vitro functional experiments demonstrated that SNHG5 overexpression significantly reduced cell apoptosis and enhanced cell viability, whereas SNHG5 knockdown in 5-Fu-resistant CRC cells increased cell apoptosis and decreased cell viability upon 5-Fu treatment. In a xenograft mouse model, we confirmed that SNHG5 overexpression led to a reduction in 5-Fu sensitivity in CRC in vivo. Mechanistically, SNHG5 acted as a molecular sponge for miR-26b. Rescue experiments validated that SNHG5 conferred 5-Fu resistance in CRC by regulating the miR-26b/Pgp axis.

CONCLUSION

SNHG5/miR-26b/Pgp regulates CRC chemosensitivity, providing potential therapeutic targets for the treatment of 5-Fu-resistant CRC.

Key Words: SNHG5; 5-fluorouracil resistance; Colorectal cancer; MiR-26b; P-glycoprotein; Long noncoding RNA; Therapeutic target

Core Tip: This study confirmed that the expression of the long noncoding RNA SNHG5 was upregulated in 5-fluorouracil (5-Fu)-resistant colorectal cancer (CRC) tissues and 5-Fu-resistant CRC cells. Functional experiments confirmed that SNHG5 overexpression reduced 5-Fu sensitivity in CRC in vitro and in vivo. Mechanistically, SNHG5 acted as a molecular sponge for miR-26b. In addition, SNHG5 conferred 5-Fu resistance in CRC by regulating the miR-26b/p-glycoprotein axis. Our results elucidate potential mechanisms of noncoding RNAs in CRC chemoresistance and provide new strategies for CRC chemotherapy.

INTRODUCTION

Colorectal cancer (CRC) is a common cancer that ranks third in incidence among human cancers and is the second leading cause of cancer-related mortality[1]. Moreover, the incidence rate of CRC in younger patients has increased over the last several decades[2]. The prognosis of CRC patients has gradually improved with the adoption of several clinical techniques, including surgery, radiation, and chemotherapy[3,4]. 5-fluorouracil (5-Fu), a first-line chemotherapeutic agent, has been widely used for CRC treatment[5]. However, resistance to 5-Fu is a critical limitation to its clinical application[6,7]. Therefore, elucidating the underlying mechanisms of 5-Fu resistance is important.

Long noncoding RNAs (lncRNAs) are classified as noncoding RNAs (ncRNAs) that are more than 200 nucleotides in length and have no protein-coding capacity. Numerous studies have shown that lncRNAs participate in 5-Fu resistance by modifying signaling pathways and regulating gene expression. For example, recent research revealed that the lncRNA NEAT1 promoted 5-Fu resistance by remodeling chromatin, which increased the acetylation levels of histones and c-Myc and promoted the expression of aldehyde dehydrogenases 1 and c-Myc[8]. The lncRNA HAND2-AS1 was shown to inhibit 5-Fu resistance by modulating the miR-20a/PDCD4 axis in CRC[9]. Additionally, the lncRNA CRNDE was shown to attenuate chemoresistance via SRSF6-regulated alternative splicing of PICALM in gastric cancer[10].

The lncRNA SNHG5 is highly expressed in a variety of tumor tissues, including CRC, and promotes tumor proliferation and invasion through multiple pathways[11-13]. However, the function and mechanism of SNHG5 in 5-Fu resistance in CRC are still unknown.

In this study, we attempted to identify the function and mechanism of SNHG5 in 5-Fu resistance in CRC. We demonstrated that the dysregulation of SNHG5 conferred 5-Fu resistance in CRC cells both in vitro and in vivo. SNHG5 modulated 5-Fu resistance by directly binding to miR-26b, which regulated the target gene p-glycoprotein (Pgp). Collectively, our findings provide new insights into the role of SNHG5 in 5-Fu resistance in CRC.

MATERIALS AND METHODS

Clinical samples

We obtained clinical colorectal tissues, including tumor tissues (n = 36) and adjacent normal tissues (n = 16), from individuals at Changshu No. 2 People’s Hospital. The clinical samples were stored in liquid nitrogen after surgical removal. The patients received postoperative adjuvant chemotherapy based on 5-Fu followed by chest X-ray, computed tomography and/or gastrointestinal endoscopy to monitor for tumor progression. Clinicopathological characteristics of the 36 CRC patients were recorded (Table 1) and the response to 5-Fu was evaluated based on the response evaluation criteria in solid tumors (version 1.1). 5-Fu resistance was defined as disease progression within 6 months after primary chemotherapy or during primary chemotherapy, whereas 5-Fu sensitivity was defined as progression beyond 6 months or no progression[14,15]. The experimental procedures involving clinical samples were approved by the Institutional Review Board of Changshu No. 2 People’s Hospital (No. 2020-KY-008).

Table 1 Clinicopathologic characteristics of 36 patients with colorectal cancer.

Characteristics	5-Fu-sensitive group (n = 22)	5-Fu-resistant group (n = 14)	P value
Age (years)
< 60	3	2	1.000
≥ 60	19	12
Sex
Male	14	9	1.000
Female	8	5
Tumor size
< 5 cm	17	9	0.462
≥ 5 cm	5	5
Location
Colon	14	6	0.307
Rectum	8	8
Differentiation
Well/moderate	15	10	1.000
Poor	7	4
Tumor depth
T1 + T2	3	2	1.000
T3 + T4	19	12
AJCC stage
II	10	4	0.485
III	12	10
SNHG5 expression
Low	17	5	0.018a
High	5	9

^aP < 0.05.

The Fisher’s exact test was applied to analyze categorical variables. 5-Fu: 5-fluorouracil; AJCC: American Joint Committee on Cancer.

Cell culture and treatment

CRC cell lines (LOVO, HT-29, HCT-8, and SW116), human normal colorectal epithelial cells (FHCs), and HEK-293T cells were obtained from American type culture collection (MD, United States), and 5-Fu-resistant cells (LOVO/5-Fu, HT-29/5-Fu) were obtained as described in our previous study[16]. CRC cell lines were cultured in Dulbecco’s modified Eagle’s medium (Gibco, Thermo Fisher Scientific, United States) supplemented with penicillin (100 U/mL) and streptomycin (100 μg/mL). Resistance cells were cultured as described previously[16].

LncRNA SNHG5 small interfering RNA (siRNA), control siRNA, miR-26b mimics and miR-control (miR-NC) mimics were obtained from RiboBio (Guangzhou, Guangdong Province, China). The pcDNA3.1-SNHG5 vector (SNHG5), pcDNA3.1-Pgp vector (Pgp) and control vectors were purchased from Biovector Co., Ltd. (Beijing, China).

Luciferase reporter assay

SNHG5 sequences with predicted miR-26b binding sites [wild-type (WT): 5’-GTGCCACGAGGTTTACTTGAC-3’] and the corresponding mutant (Mut) sequences (Mut: 5’-GTGCCACGAGGTAATGAACTC-3’) were cloned and inserted into the pmirGLO vector (Promega, Madison, WI, United States). Before the luciferase reporter assay, HEK-293T cells were seeded and cultured in 24-well plates. The plasmids encoding SNHG5-WT or SNHG5-Mut (500 ng) were transfected with miR-NC or miR-26b mimics (100 nM) into HEK-293T cells. After 24 hours, the cells were lysed with 1 × passive lysis buffer for 15 minutes, and a dual-luciferase reporter assay system was used to measure luciferase activity (E1910, Promega, Madison, WI, United States).

RNA extraction and quantitative real-time polymerase chain reaction

Total RNA was extracted using TRIzol reagent (15596026, Invitrogen, Carlsbad, CA, United States). RNA was isolated from the cytoplasm and nucleus of CRC cells using a cytoplasmic and nuclear RNA purification kit (Cat. 21000, 37400, Norgen, Thorold, Canada). Complementary DNA (cDNA) for lncRNA analysis was synthesized using a lnRcute lncRNA cDNA kit (KR202-02, TIANGEN, Beijing, China), and cDNA for miRNA analysis was synthesized using a HiScript cDNA synthesis kit (R323-01, Vazyme, Nanjing, China). Beta-actin (β-actin) and U6 were used as controls for the mRNA and miRNA assays, respectively. Quantitative polymerase chain reaction (qPCR) was performed using ChamQ SYBR qPCR master mix (high ROX premixed) (Q341-02, Vazyme, Nanjing, Jiangsu Province, China) according to the manufacturer’s instructions. The sequences of primers used were as follows: SNHG5 (forward) 5’-TACTGGCTGCGCACTTCG-3’; SNHG5 (reverse) 5’-TACCCTGCACAAACCCGAAA-3’; miR-26b (forward) 5’-ACACTCCAGCTGGGTTCAAGTAATTCAGG-3’; miR-26b (reverse) 5’- CTCAACTGGTGTCGTGGA-3’; β-actin (forward) 5’-CATGTACGTTGCTATCCAGGC-3’; β-actin (reverse) 5’-CTCCTTAATGTCACGCACGAT-3’; U6 (forward) 5’-CTCGCTTCGGCAGCACA-3’; and U6 (reverse) 5’-AACGCTTCACGAATTTGCGT-3’.

RNA immunoprecipitation assay

LOVO and HT-29 cells transfected with miR-26b were scraped from culture plates, resuspended in ice-cold PBS, and then lysed in RNA immunoprecipitation (RIP) lysis buffer [25 mmol/L tris potential of hydrogen (pH) = 7.4, 150 mmol/L potassium chloride, 5 mmol/L ethylene diamine tetraacetic acid, 0.5 mmol/L dithiothreitol, 0.5% nonaethylene glycol octylphenyl ether, 100 U/mL RNAase inhibitor SUPERase, and protease inhibitors] (Abcam, United States). The lysates were incubated with a human anti-argonaute 2 (Ago2) antibody (67934-1-Ig, Proteintech, IL, United States) overnight [a human anti-IgG antibody (10284-1-AP, Proteintech, IL, United States) was used as a control] and then incubated with protein A/G beads for one hour. After the beads were washed twice, RNA was isolated with TRIzol reagent and analyzed via quantitative real-time polymerase chain reaction (qRT-PCR).

Stable cell line construction

SNHG5-expressing (SNHG5) and scrambled control, SNHG5-short hairpin RNA (shRNA) (shSNHG5) and negative control (shNC) lentiviruses were purchased from GenePharma (Shanghai, China). After infection, stable cells were selected with puromycin (2 μg/mL; Roche, United States) for 2 weeks. Stable cells were identified via qRT-PCR and pooled for subsequent analysis.

Cell viability and apoptosis assays

SNHG5-overexpressing CRC cells and SNHG5-knockdown 5-Fu-resistant cells were cultured in 96-well plates (5000 cells/well). After 24 hours, the cells were treated with different concentrations of 5-Fu for 48 hours, and then, cell viability was analyzed using 10% cell counting kit 8 (CCK-8) (Dojindo, Kumamoto, Japan). Stable cells were treated with 2.5 μg/mL 5-Fu for 48 hours. Flow cytometry was performed after annexin-V fluorescein 5-isothiocyanate/propidium iodide staining (A211-01, Vazyme, Nanjing, Jiangsu Province, China) according to the manufacturer’s instructions.

Western blotting

Total protein was isolated from cells using cell lysis buffer [20 mmol/L tris (pH = 7.5), 150 mmol/L sodium chloride, 1% Triton X-100, and phenyl methane sulfonyl fluoride]. Then, 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels were used to separate the proteins. Primary antibodies against Pgp (1:1000; ab261736, Abcam, United States), Bax (1:1000; 50599-2-Ig, Proteintech, IL, United States), Bcl-2 (1:1000; 26593-1-AP, Proteintech, IL, United States) and β-actin (1:1000; 4967S, Cell Signaling Technology, United States) were used.

Tumorigenesis in nude mice and immunohistochemistry

Stable cells [SNHG5-overexpressing LOVO (LOVO/SNHG5), control LOVO (LOVO/Con), SNHG5-knockdown LOVO/5-Fu (LOVO/5-Fu/shSNHG5), and control LOVO/5-Fu (LOVO/5-Fu/shNC) cells] (5 × 10⁶ cells per mouse) were injected subcutaneously into 4-week-old BALB/c-nude mice. Approximately 7 days later, 30 mg/kg 5-Fu in saline buffer was administered by intraperitoneal injection. The mice were treated with a cyclic regimen composed of three subsequent daily injections followed by two days of recovery. Tumors were measured every two days. After two weeks, the tumors were isolated from the mice and weighed prior to subsequent analyses. Tumor volume was calculated using the following formula: V = 1/2 a² × b, where a is tumor length and b is tumor width. Immunohistochemistry (IHC) was performed using an anti-Ki-67 antibody (ab15580, Abcam, United States) to analyze the proliferation of tumors in different groups. IHC experiments were performed in accordance with the manufacturer’s instructions (Genetech, China, GK500705). The Ki-67 positive area (%) was calculated by Image J software (NIH, United States). The animal experiments were approved by the Animal Care and Welfare Committee of Southeast University (No. 20200721007).

Statistical analysis

Continuous data are expressed as the mean ± SD. The Fisher’s exact test was used to analyze categorical variables. For continuous variables, Student’s t test was used to analyze differences between two groups, and one-way analysis of variance was used to determine significant differences among three or more groups. Correlations were analyzed via the Pearson correlation coefficient. Data were considered statistically significant when the P value was < 0.05.

RESULTS

Upregulated SNHG5 expression is associated with 5-Fu resistance in CRC

We first explored the expression levels of SNHG5 in CRC tissues (n = 36) and normal colorectal tissues (n = 16) via qRT-PCR. Our results revealed that the level of SNHG5 in CRC tissues was nearly 3.64-fold greater than that in normal tissues (P < 0.01; Figure 1A). We subsequently examined the expression of SNHG5 in CRC by analyzing data from the Cancer Genome Atlas database. We compared data from 622 CRC tissues and 51 normal colorectal tissues and found that the expression level of SNHG5 was 1.77-fold higher in CRC tissues than in normal colorectal tissues (P < 0.01; Figure 1B). We measured the expression levels of SNHG5 in the LOVO, HT-29, HCT-8 and SW116 CRC cell lines and in a normal colorectal epithelial cell line, i.e., FHC cells. We found that SNHG5 expression was higher in the CRC cell lines than in the FHC cell line (P < 0.01; Figure 1C). These results indicated that SNHG5 was highly expressed in CRC.

Figure 1 Upregulated SNHG5 expression is associated with 5-fluorouracil resistance in colorectal cancer. A and B: The expression levels of SNHG5 were assessed in collected samples [normal tissues = 16, colorectal cancer (CRC) tissues = 36] and The Cancer Genome Atlas database; C: The expression of SNHG5 was evaluated by quantitative real-time polymerase chain reaction (qRT-PCR) in CRC cell lines (LOVO, HT-29, HCT-8, and SW116) compared to normal colorectal epithelial cell line; D: The expression levels of SNHG5 were examined in the 5-fluorouracil (5-Fu)-sensitive and 5-Fu-resistant CRC groups; E and F: SNHG5 expression levels in HT-29 and HT-29/5-Fu or LOVO and LOVO/5-Fu cells were assessed by qRT-PCR. ^bP < 0.01. 5-Fu: 5-fluorouracil; CRC: Colorectal cancer.

Next, we divided CRC patients into a 5-Fu-sensitive group (n = 22) and a 5-Fu-resistant group (n = 14) and found that SNHG5 expression was significantly higher in the tumor tissues of the 5-Fu-resistant group than in those of the 5-Fu-sensitive group (1.77-fold, P < 0.01; Figure 1D). The association between chemoresponse and SNHG5 expression was also assessed by analyzing 36 samples (Table 1). The results revealed a significant correlation between 5-Fu resistance and high expression levels of SNHG5 in CRC (P < 0.05). Next, we examined SNHG5 expression in CRC cell lines and constructed 5-Fu-resistant cell lines. As shown in Figure 1E and F, 5-Fu-resistant cell lines expressed higher levels of SNHG5 than parent CRC cells did. These results suggested that SNHG5 may play an important role in 5-Fu resistance in CRC.

SNHG5 enhances 5-Fu resistance in CRC cells

To investigate the effect of SNHG5 on the chemosensitivity of CRC cells to 5-Fu, we first transfected an SNHG5-expressing lentivirus (SNHG5) and a control vector into LOVO and HT-29 cells to overexpress SNHG5 (Figure 2A). CCK-8 assay results revealed that LOVO/SNHG5 and HT-29/SNHG5 CRC cells presented significantly decreased 5-Fu sensitivity (Figure 2B). The calculation of the IC₅₀ values for SNHG5 and control in LOVO cells (19.97 ± 0.88 μM vs 6.67 ± 0.60 μM; P < 0.01) and HT-29 cells (22.15 ± 1.10 μM vs 4.85 ± 0.25 μM; P < 0.01) verified the ability of SNHG5 to decrease 5-Fu sensitivity (Figure 2C). Apoptosis assays revealed that the apoptotic rate of SNHG5-overexpressing CRC cells treated with 5-Fu was significantly lower than that of control cells (P < 0.01; Figure 2D). Consistent with the results shown in Figure 2D, SNHG5 overexpression decreased Bax levels and increased Bcl-2 levels in the presence of 5-Fu (Figure 2E). These results suggested that the overexpression of SNHG5 contributed to 5-Fu resistance in LOVO and HT-29 cells.

Figure 2 SNHG5 overexpression enhances 5-fluorouracil resistance in colorectal cancer cells. A: The expression of SNHG5 was assessed in LOVO and HT-29 cells which were transfected with SNHG5 and scrambled control lentivirus; B and C: The viability of LOVO and HT-29 cells treated with different concentrations of 5-fluorouracil (5-Fu) was evaluated, and the IC50 of 5-Fu in colorectal cancer cells was calculated; D: The apoptosis rate of stable cells treated with 5-Fu was analyzed via annexin-V fluorescein 5-isothiocyanate/propidium iodide staining assays (left) and quantified (right); E: Western blot analysis was performed to assess the expression of apoptosis markers (Bax and Bcl-2), with β-actin serving as an internal control (left), and densitometric analysis of the Western blot signals is shown (right). ^bP < 0.01. Con: Control; 5-Fu: 5-fluorouracil.

Next, we investigated the effect of SNHG5 knockdown on 5-Fu sensitivity via the use of an SNHG5-shRNA lentivirus (shSNHG5) (Figure 3A). CCK-8 assay results revealed that the knockdown of SNHG5 significantly decreased the 5-Fu resistance of 5-Fu-resistant CRC cells (Figure 3B). We measured the IC₅₀ and found that SNHG5-knockdown cells were more sensitive to 5-Fu than control cells (7.40 ± 0.57 μM vs 20.56 ± 1.92 μM for LOVO, P < 0.001; 7.35 ± 0.89 μM vs 27.37 ± 4.85 μM for HT-29, P < 0.001) (Figure 3C). In contrast to SNHG5 overexpression, SNHG5 knockdown increased cell apoptosis and Bax levels and decreased Bcl-2 levels in CRC cells (Figure 3D and E).

Figure 3 SNHG5 knockdown suppresses 5-fluorouracil resistance in 5-fluorouracil-resistant colorectal cancer cells. A: The expression of SNHG5 was assessed in 5-fluorouracil (5-Fu)-resistant colorectal cancer (CRC) cells (LOVO/5-Fu, HT-29/5-Fu) which were transfected with SNHG5-short hairpin RNA and negative control lentivirus; B and C: The viability of LOVO/5-Fu and HT-29/5-Fu cells treated with 5-Fu at different concentrations was evaluated, and the IC₅₀ of 5-Fu-resistant CRC cells was calculated; D: The apoptosis rate of stable cells treated with 5-Fu was analyzed using annexin-V fluorescein 5-isothiocyanate/propidium iodide staining assays (left) and quantified (right); E: Western blot analysis was performed to assess the expression of markers of apoptosis (Bax and Bcl-2), with β-actin as an internal control (left), and densitometric analysis of Western blot signals is shown (right). ^bP < 0.01. ^cP < 0.001. shSNHG5: SNHG5-short hairpin RNA; shNC: Negative control-short hairpin RNA; 5-Fu: 5-fluorouracil.

SNHG5 promotes 5-Fu resistance in CRC in vivo

To evaluate whether SNHG5 promotes 5-Fu resistance in CRC in vivo, LOVO/SNHG5 cells, LOVO/Con cells, LOVO/5-Fu/shSNHG5 cells and LOVO/5-Fu/shNC cells were injected subcutaneously into the flanks of 4-week-old nude mice. When the tumor diameter reached 5 mm, 5-Fu was administered by intraperitoneal injection. The results revealed that the tumors derived from LOVO/SNHG5 cells in 5-Fu-treated mice had greater mean volumes, exhibited faster growth, and were heavier than those derived from LOVO/Con cells. In contrast, the tumors derived from LOVO/5-Fu/shSNHG5 cells presented smaller mean volumes, exhibited slower growth, and were lighter than those derived from the LOVO/5-Fu/shNC cells (Figure 4A-C). The proliferation marker Ki-67 was used to test tumorigenesis ability in vivo. Our results revealed that the LOVO/SNHG5 group had a greater Ki-67-positive than the LOVO/Con group did; however, the LOVO/5-Fu/shSNHG5 group presented a lower positive rate than the LOVO/5-Fu/shNC group did (Figure 4D). Taken together, these data indicated that SNHG5 promoted CRC resistance to 5-Fu in vivo.

Figure 4 SNHG5 promotes 5-fluorouracil resistance in colorectal cancer in vivo. A: Tumor growth after treatment with 5-fluorouracil was assessed using the following formula: V = 1/2 × a² × b, where a is tumor length and b is tumor width; B: Representative images of trimmed tumors; C: Tumor weight was calculated; D: Ki-67 expression was evaluated in tumor tissues by immunohistochemistry (scale bar = 400 μm). ^aP < 0.05. ^bP < 0.01. Con: Control; shSNHG5: SNHG5-short hairpin RNA; shNC: Negative control-short hairpin RNA; 5-Fu: 5-fluorouracil.

SNHG5 promotes CRC cell resistance to 5-Fu by binding to miR-26b

To identify the mechanism by which SNHG5 promotes 5-Fu resistance, we first examined, via qRT-PCR analyses, the cellular location of SNHG5 and found that SNHG5 was enriched in both the cytoplasmic and nuclear fractions of LOVO and HT-29 cells (Figure 5A). Multiple studies have shown that lncRNAs can function as competing endogenous RNAs (ceRNAs) by binding to miRNAs[17,18]. We subsequently searched for miRNAs that have complementary base pairing with SNHG5 via the online software programs miRcode (http://www.miRcode.org) and DIANA-LNCBase (https://diana.e-ce.uth.gr/Lncbasev3). We previously reported that miR-26b, which has been found to suppress 5-Fu resistance by inhibiting Pgp in CRC[16], may form complementary base pairs with SNHG5 (Figure 5B).

Figure 5 SNHG5 directly binds to miR-26b in colorectal cancer. A: Quantitative real-time polymerase chain reaction (qRT-PCR) was used to analyze SNHG5 expression in the nuclear and cytoplasmic fractions of LOVO and HT-29 cells; B: A Venn diagram was used to identify potential miRNAs that may target the binding region of SNHG5; C: The expression levels of miR-26b were evaluated via qRT-PCR in LOVO and HT-29 cells transfected with SNHG5 and control or long noncoding RNAs SNHG5 small interfering RNA (siRNA), control siRNA; D: A dual-luciferase assay was used to assess the seed-matching sites or mutation sites between SNHG5 and miR-26b in HEK-293T cells; E: Anti-argonaute 2 RNA immunoprecipitation analyses were performed using LOVO and HT-29 cells transfected with miR-26b, and the enrichment of SNHG5 was analyzed via qRT-PCR. ^bP < 0.01. Si-NC: Control small interfering RNA; Si-SNHG5: SNHG5 small interfering RNA; WT: Wild type; Mut: Mutant; miR-NC: MiR-control; NS: No significance; CMV: Cytomegalovirus; Ago-2: Argonaute 2.

To confirm the interaction between SNHG5 and miR-26b, we transfected LOVO and HT-29 cells with an SNHG5-overexpression vector or siRNA and found that miR-26b expression was significantly downregulated or upregulated, respectively (Figure 5C). We subsequently constructed SNHG5 Luciferase reporter plasmids containing WT (SNHG5-WT) and Mut-binding sites (SNHG5-Mut) for miR-26b and transfected them into HEK-293T cells together with a miR-26b mimic (miR-26b) or a control mimic (miR-NC). The miR-26b mimic reduced the luciferase activity of the WT but not the Mut plasmid (Figure 5D). We then conducted anti-Ago2 RIP using LOVO and HT-29 cells transiently overexpressing miR-26b. Endogenous SNHG5 pulled down by Ago2 was significantly enriched in miR-26b-transfected cells (Figure 5E). These data suggested that miR-26b was a miRNA binding partner of SNHG5.

To clarify whether miR-26b is involved in SNHG5-mediated 5-Fu resistance in CRC cells, we increased miR-26b expression with a miR-26b mimic in SNHG5-overexpressing CRC cells and in SNHG5-knockdown 5-Fu-resistant CRC cells. Cell viability analyses revealed that SNHG5 promoted CRC cell proliferation and that this effect was inhibited by the miR-26 mimic. Similarly, the miR-26b mimic inhibited the proliferation of SNHG5-knockdown 5-Fu-resistant CRC cells (Figure 6A). Consistently, the miR-26b mimic increased 5-Fu-induced apoptosis and Bax expression and decreased Bcl-2 expression in SNHG5-overexpressing CRC cells and in SNHG5-knockdown 5-Fu-resistant CRC cells (Figure 6B and C).

Figure 6 SNHG5 promotes colorectal cell resistance to 5-fluorouracil by sponging miR-26b. The stable cells were transfected with miR-26b or miR-control (miR-NC), after 24 hours, and then treated with 5-fluorouracil. A: Cell viability was analyzed using cell counting kit 8 assays; B: The apoptosis rate of stable cells was analyzed using annexin-V fluorescein 5-isothiocyanate/propidium iodide staining assays (left) and quantified (right); C: Western blot analysis was performed to assess the expression of markers of apoptosis (Bax and Bcl-2), with β-actin serving as an internal control (left), and a densitometric analysis of Western blot signals is shown (right). ^bP < 0.01. ¹P compared with the control + miR-NC group or the negative control-short hairpin RNA (shNC) + miR-NC group. ²P compared with the control + miR-26b group or the shNC + miR-26b group. ³P compared with the SNHG5 + miR-NC group or the SNHG5-short hairpin RNA + miR-NC group. miR-NC: MiR-control; Con: Control; shSNHG5: SNHG5-short hairpin RNA; shNC: Negative control-short hairpin RNA; 5-Fu: 5-fluorouracil.

Taken together, these data confirmed that SNHG5 promoted CRC cell resistance to 5-Fu by binding to miR-26b.

SNHG5 contributes to 5-Fu resistance in part by regulating the miR-26b/Pgp axis

Since our previous study showed that miR-26b suppressed 5-Fu resistance by targeting Pgp[16], we investigated whether SNHG5 regulated the miR-26b/Pgp axis. We first evaluated the expression levels of miR-26b and Pgp in 36 CRC tissues. The results indicated that miR-26 expression was downregulated whereas Pgp expression was upregulated in tumor tissues of 5-Fu-resistant group compared with 5-Fu-sensitive group (Figure 7A and B). Next, we measured the correlation of SNHG5 with miR-26b and Pgp in CRC tissues. The results revealed that SNHG5 was negatively correlated with miR-26b (Figure 7C) and positively correlated with Pgp, the target of miR-26b (Figure 7D). These data indicated that SNHG5 played a vital role in CRC chemoresistance. The Western blot results revealed that SNHG5 increased the expression of Pgp; however, the miR-26b mimic reversed that effect (Figure 7E). The knockdown of SNHG5 suppressed Pgp and Bcl-2 expression and increased Bax expression, whereas Pgp overexpression partially restored the function of SNHG5 (Figure 7F). Collectively, the results demonstrated that SNHG5 served as a miR-26b sponge and increased the expression of Pgp, the target of miR-26b in CRC cells, indicating that SNHG5 conferred 5-Fu resistance in CRC at least partly by regulating the miR-26b/Pgp axis.

Figure 7 SNHG5 contributes to 5-fluorouracil resistance in part by regulating the miR-26b/p-glycoprotein axis. A and B: The expression levels of miR-26b and p-glycoprotein (Pgp) were evaluated in the colorectal cancer tissues; C and D: Pearson’s correlation analysis was performed to analyze the correlation between SNHG5 and miR-26b, SNHG5 and Pgp; E: Western blotting was performed to assess the expression levels of Pgp in LOVO and HT-29 cells transfected with control + miR-control (miR-NC), SNHG5 + miR-NC, control + miR-26b, or SNHG5 + miR-26b. Densitometric analysis of the Western blot signals is shown; F: Western blotting was performed to assess the expression levels of Pgp, Bax and Bcl-2 in LOVO cells transfected with negative control-short hairpin RNA (shNC) + vector, SNHG5-short hairpin RNA (shSNHG5) + Vector, shNC + Pgp, or shSNHG5 + Pgp. Densitometric analysis of the Western blot signals is shown. ^aP < 0.05. ^bP < 0.01. ^cP < 0.001. miR-NC: MiR-control; Con: Control; shSNHG5: SNHG5-short hairpin RNA; shNC: Negative control-short hairpin RNA; 5-Fu: 5-fluorouracil; Pgp: P-glycoprotein.

DISCUSSION

As a common clinical therapy, chemotherapy, especially 5-Fu-based chemotherapy, is still the main therapeutic option for patients with CRC at an advanced stage or at high risk of recurrence[19]. Increasing evidence has demonstrated that the phenomenon of drug resistance may become a major obstacle to successful outcomes during clinical adjuvant chemotherapy[20]. Recently, mechanisms of drug resistance in cancer, such as epigenetic alterations and genetic dysregulation, have been revealed[21,22]. However, the mechanisms of drug resistance in CRC are complex, involve multiple factors and still need to be investigated. This study attempted to identify the function and mechanism of SNHG5 in 5-Fu resistance in CRC. We confirmed that SNHG5 expression was upregulated in 5-Fu-resistant CRC tissues and 5-Fu-resistant CRC cells. SNHG5 overexpression led to a reduction in 5-Fu sensitivity in CRC in vitro and in vivo. Mechanistically, SNHG5 acted as a molecular sponge for miR-26b. SNHG5 conferred 5-Fu resistance in CRC by regulating the miR-26b/Pgp axis. In conclusion, our study innovatively demonstrates SNHG5/miR-26b/Pgp axis regulates CRC chemosensitivity, providing potential therapeutic targets for the treatment of 5-Fu-resistant CRC.

A review of recent literature indicates that SNHG5 plays a vital role in human cancers[23]. Chi et al[24] reported that SNHG5 facilitated the proliferation of breast cancer cells by sponging the miR-154-5p/proliferating cell nuclear antigen axis. Zhang et al[11] reported that in CRC, SNHG5 affected cell proliferation, metastasis, and migration by regulating miR-132-3p/CREB5. Damas et al[25] reported that SNHG5 promoted CRC cell survival by counteracting STAU1-mediated mRNA destabilization. However, the functions of SNHG5 in promoting 5-Fu resistance in CRC are currently unclear. In the present study, we found that higher expression levels of SNHG5 were associated with 5-Fu resistance in CRC. Our data demonstrated that SNHG5 overexpression significantly reduced cell apoptosis and enhanced cell viability, whereas SNHG5 knockdown in 5-Fu-resistant CRC cells increased cell apoptosis and decreased cell viability upon 5-Fu treatment. In a xenograft mouse model, we observed that 5-Fu, in combination with shSNHG5, inhibited the growth of 5-Fu-resistant CRC cells, thus confirming that SNHG5 promoted the resistance of CRC cells to 5-FU both in vitro and in vivo.

Multiple studies have shown that SNHG5 regulates chemoresistance in malignancies via various mechanisms, including ceRNAs. Wang et al[26] confirmed that SNHG5 regulated the resistance of acute myeloid leukemia to chemotherapy by targeting the miR-32/DNAJB9 axis. Li et al[27] reported that the lncRNA SNHG5 promoted cisplatin resistance in gastric cancer by regulating drug resistance-related genes. To elucidate the mechanism by which SNHG5 promotes 5-FU resistance, we predicted potential miRNAs that may bind with SNHG5 using DIANA and miRcode and identified miR-26b as a predicted binding partner of SNHG5. Luciferase reporter and RIP assay results verified the binding of SNHG5 to miR-26b. Further experiments confirmed that miR-26b reversed the results of 5-Fu resistance, a low apoptotic rate and a high level of Pgp induced by overexpressing SNHG5. However, miR-26b further increased the apoptosis of SNHG5-knockdown 5-FU-resistant CRC cells. In general, we found that SNHG5 promoted 5-Fu resistance in CRC cells via the miR-26b/Pgp axis.

The adenosine triphosphate (ATP) binding cassette (ABC) superfamily, a group of membrane transporter proteins, includes 7 subfamilies, ABC-A to ABC-G, in humans and can function as ATP-dependent efflux pumps of antitumor drugs[28]. Pgp, also called ABCB1, is the first identified ABC transporter and contribute to chemotherapy failure in various cancer cells, including CRC cells[29]. Pgp can reduce the accumulation of intracellular antitumor drugs through the export of drugs, leading to multidrug resistance in CRC[30]. The substrates of Pgp include commonly utilized chemotherapeutic agents, which subsequently results in intrinsic resistance in CRC cells. Furthermore, exposure to these chemotherapeutic drugs induces the upregulation of Pgp in CRC cells, leading to acquired resistance[31]. Wang et al[29] reported that Pgp was highly expressed in CRC tissues and correlated with poor prognosis in CRC patients. We found that the expression of Pgp was higher in 5-Fu-resistant CRC tissues than in 5-Fu-sensitive tissues. However, the difference between the two groups was not statistically significant due to the limited sample size. The regulatory networks of Pgp remain unclear, and the potential regulatory mechanisms of ncRNAs have gained much attention[32]. Shi et al[33] reported that miR-29a overexpression restored the sensitivity of doxorubicin (DOX)-resistant CRC cells to DOX treatment via the downregulation of Pgp. Our previous data confirmed the tumor suppressive function of miR-26b in increasing the sensitivity of CRC cells to 5-Fu by negatively regulating Pgp expression[16]. As a further extension, the present study revealed that SNHG5 served as a miR-26b sponge and increased the expression of Pgp in CRC cells. These findings verified the regulatory role of the lncRNA SNHG5/miR-26b/Pgp axis in 5-Fu resistance in CRC and clarified the underlying mechanism.

This study also has several limitations. Firstly, our study focused on the regulation of the SNHG5/miR-26b/Pgp axis, which might be a subset of a vast regulatory network in 5-Fu resistance in CRC, and other factors affecting 5-Fu resistance should be explored further. Secondly, the findings were investigated based on in vitro and xenograft mouse models, which might not recapitulate the complexity of CRC. With advances in gene editing techniques, transgenic animal models are applied to validate molecular functions, which are more consistent with the clinical pathogenic mechanism. It is valuable to utilize transgenic animal models with overexpression or specific knockout of SNHG5 to validate the role of SNHG5 in the chemoresistance of CRC in subsequent studies. Additionally, we detected the expression of SNHG5/miR-26b/Pgp in CRC tissues with a limited number of samples. It is necessary to further validate the findings in a large sample size cohort and detect the expression of SNHG5/miR-26b/Pgp in serum to explore novel non-invasive biomarkers for CRC chemotherapy.

CONCLUSION

Overall, in our study, we confirmed that the lncRNA SNHG5 promoted drug resistance in 5-Fu-resistant CRC. SNHG5 suppressed the sensitivity of CRC cells to 5-Fu treatment and facilitated the growth of cancer cells via miR-26b/Pgp. Our results contribute to elucidating the potential mechanisms of ncRNAs in the chemoresistance of CRC and provide new strategies for CRC chemotherapy.

ACKNOWLEDGEMENTS

The authors acknowledge Zhou GQ, Shi ZL, Guo J, Lu FY, Zhu JL and Wei W for their valuable assistance.