The original CRC gene expression profile data and related clinical data of six independent gene chips GSE21510, GSE22598, GSE23878, GSE32323, and GES39582 were obtained from the GEO database. GSE21510 contains 104 tumor and 44 normal tissues, GSE22598 contains 17 tumor and 17 normal tissues, GSE23878 contains 35 tumor and 24 normal tissues, GSE32323 contains 17 tumor and 17 normal tissues, and GES39582 contains 233 tumor and 19 normal tissues. Significant Analysis of Microarray software was used to analyze the expression differences of lncRNAs in normal colorectal mucosal and CRC tissue samples from the two groups ( Supplementary Table 2 ).

Tumor tissues paired with normal colorectal mucosal tissues were obtained from 36 patients with CRC diagnosed and operated on at the Hunan Cancer Hospital from 2017 to 2019. After collection, the tissue specimens were immediately frozen in liquid nitrogen. In addition, in situ hybridization assays were performed on 53 patients who underwent surgery at the Human Provincial Cancer Hospital from 2010 to 2013. 10 normal samples were obtained from adjacent tissues. All specimens were obtained from surgically removed tissue specimens from patients with preoperatively confirmed CRC who had not received radiation or chemotherapy. This study was approved by the ethics committee of the Hunan Cancer Hospital, and written informed consent was obtained from all patients.

Tissue samples from 53 patients with CRC were fixed with 4% paraformaldehyde, embedded in paraffin, and sliced. The expression of DNAJC3-AS1 in CRC tissues was measured with an oligonucleotide probe. The following oligonucleotide probes were purchased from BOSTER Company (Wuhan, China): 5′-TGAAT TATAA ATAGC ATAGT GAATT TGTGA TTCCC TGAAG-3′, 5′-TCCCT CCTTC CTTGA GTGTG GGCTG GACTT AGTGA CTAAC-3′, 5′-TTGGG ACCTC CCTGT ATTAT CCTTA TGCCC TTACT TGAAA-3′, and 5′-labeled with a digoxigenin-deoxyuridine triphosphate tag. The experiment was conducted according to the manufacturer’s protocol using a sensitive, enhanced ISH kit (Boster, Wuhan, China). A semi-quantitative scoring standard was used to record the staining intensity and positive area of the slice. The scoring was graded as 0 (negative), 1 (<10% positive), 2 (10%–50% positive), or 3 (>50% positive) in accordance with the staining proportion and intensity. The final scores were regarded as low expression (0–1) and high expression (2–3). All sections were evaluated blindly by two pathologists.

Two pathologists scored the results according to the following criteria: (1) staining intensity: no observed cell staining was scored as 0; cells with light-brown cell staining as 1; cells stained brown with no background staining, or dark-brown stained cells with light-brown background staining were recorded as moderately positive, as 2; dark-brown stained cells with no background staining were recorded as strongly positive, as 3. (2) number of positive cells: no positive cells were scored as 0; less than 25% of positive cells as 1; between 25 and 50% positive cells as 2; positive cells over 50% recorded as strongly positive, as 3. The final scores were obtained by multiplying the two scores. The results were as follows: 0 was considered as negative expression, and the final score was 0; 1−2, as weakly positive, and the final score was 1; 3 to 4, as moderately positive, and the final score was 2; 6 to 9 as strong positive, and the final score was 3.

All the target cells of each tissue were counted under 10× magnification, and the counting was repeated twice. The average values obtained from the counting was used for statistical analysis. ISH scores of 1, 2, and 3 indicated high expression of DNAJC3-AS1, and IHC scores of 0 indicated low expression of DNAJC3-AS1. All the samples were scored independently by two experienced pathologists who were double-blinded.

Total RNA was extracted using TRIzol reagent according to the manufacturer’s protocol (Invitrogen, USA). cDNA was synthesized using a reverse transcription kit (Bio-Rad, Hercules, CA, USA). qRT-PCR was performed using SYBR Green (Bio-rad, Hercules, CA, USA) with a LightCycler 480 RT-PCR detection system (Roche). β-actin was used as an internal control for normalization. The expression of DNAJC3-AS1 was normalized to the corresponding β-actin expression level. The following equation was used to calculate the relative expression: ΔCt = Ct (target gene) − Ct (β-actin), fold expression = 2 – [ΔCt (tumor) − ΔCt (normal)] divided by the Cq value. The primer sequences used in the experiment were as follows: DNAJC3-AS1: 5′-AGCGATTGTGGAAGACCCTG-3′, 5′-ATTTCCCCTGGTAAGCGCAA-3′ and β-actin: 5′-TCACCAACTGGGACGACATG-3′, 5′-GTCACCGGAGTCCATCACGAT-3′.

The CRC cell lines HT29, HCT116, SW480, and SW620 and normal human colon tissue cells CCD-18Co were obtained from the Cancer Institute of Xiangya Medical College, Central South University. The CRC cell lines HT29 and SW620 were cultured in RPMI-1640 medium (BI) with 10% foetal bovine serum (FBS, ZETA) and maintained at 5% CO₂ and 37°C. The colorectal cell line CCD-18CO and the CRC cell lines HCT116 and SW480 were cultured in DMEM (Gibco, Beijing, China) with 10% FBS (ZETA).

A small interfering RNA (siRNA) was used to interfere with the expression of DNAJC3-AS1 in cells. Cells were seeded in a 12-well plate in Opti-MEM medium (Invitrogen) and transfected with an siRNA targeting DNAJC3-AS1 or a scrambled control siRNA using Hiperfect (Takara). The sequences of the siRNA targeting DNAJC3-AS1 were as follows: siDNAJC3-AS1: sense 5′- GGAAGCACAGUCUCAACUUTT-3′ and antisense: 5′- AAGUUGAGACUGUGCUUCCTT-3′. The sequences of the scrambled control siRNA were provided by Sangon Biotech Company (Shanghai, China).

The proliferation ability of the cells was evaluated by CCK-8 and colony formation assays. In the CCK-8 assay, 100 µl of suspension containing 1 × 10³ CRC cells was inoculated into each 96-well plate. Ten microliters of CCK-8 solution was added to each well, and the cells were incubated at 37°C for 4 h. The absorbance at 450 nm was measured at the indicated time points (24, 48, 72, and 96 h after transfection).

In the colony formation assay, 2 × 10³ CRC cells per well were seeded in a 6-well plate, cultured in 2 mL DMEM containing 10% FBS and 2 mL of RPMI 1640 medium containing 10% FBS, and incubated at 37°C and 5% CO₂ for 24 h. The medium was changed every 2 days, and the culture was terminated on the 12^th day. The cells were immobilized with 4% paraformaldehyde for 30 min, purified twice with phosphate-buffered saline (PBS), and stained with 2.5% crystal violet. Subsequently, live colonies were counted.

The migration ability of CRC cells was measured by wound healing assays. The cells were seeded in a 6-well plate and grown until 70% confluence. Vertical scratches in the cell monolayer were created using a 10-µl pipette tip, and the cells were washed three times with PBS to remove cell debris. The width of the wound was measured under a microscope at 0, 24, and 48 h after the scratch wound formation, and the wound area was calculated using Image J.

The invasion ability of CRC cells was evaluated by the Matrigel Transwell invasion assay. A total of 2 × 10⁵ cells in 200 µl serum-free medium were added to the top of a Matrigel-coated Transwell cell culture chamber (8-μm pore size, BD Biosciences, New Jersey, USA), and cells in 600 µl of medium containing 20% FBS were added to the bottom of the chamber. Cells were then incubated at 37°C for 24 h. The cells that migrated or invaded the lower cavity were fixed with 4% paraformaldehyde and stained with 2.5% crystal violet. The cells on the upper surface of the chamber were gently wiped with a cotton swab. The number of invasive cells was counted from six randomly selected 100× fields under a microscope and displayed as the average value of each field.

To visualize lipid droplets, cultured cells were fixed in 4% paraformaldehyde solution on the 6-well plates, stained with 0.05 μg/ml Nile red (Solarbio) for 10 min, washed with PBS, then stained with DAPI. The images were visualized by immunofluorescence microscopy.

Total protein was extracted from the cells using RIPA buffer (Boster, China) containing a phosphatase inhibitor and quantified using the bicinchoninic acid protein detection kit (Invitrogen, USA) according to the manufacturer’s instructions. After denaturation, 20 mg of protein samples were loaded into a sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel for electrophoresis. After transfer of protein bands, the polyvinylidene fluoride membrane (Invitrogen, USA) was blocked with bovine serum albumin (Invitrogen, USA) at 25°C for 1 h and incubated with the following primary antibodies overnight at 4°C: anti-phospho-EGFR (1:1,000, Abclone), anti-phosphorylated PI3K (1:1,000, Abclone), anti-phosphorylated AKT (1:1,000, Abclone), anti-phosphorylated NF-κB (1:1,000, Abclone), anti-SREBP1 (1:1,000, Abcam), anti-ACC1 (1:1000, Proteintech), anti-FASN (1:1,000, Proteintech), and β-actin (1:1,000, Abclone). After washing with Tris-buffered saline with 0.1% Tween^® 20 (TBS-T), the membrane was incubated with an anti-rabbit secondary antibody for 2 h at room temperature. After another washing with TBS-T, protein bands were visualized using an enhanced chemiluminescent substrate kit (Obsen, Beijing). Experiments were performed in triplicate.

All assays were performed independently at least three times. All statistical analyses were performed using Microsoft Excel 2007 version (Microsoft, USA) and GraphPad Prism version 5.0 (GraphPad Software Inc., San Diego, CA, USA) software. Data are expressed as the mean ± standard error of the mean. Differences between two independent groups were assessed by Student’s t-test, and the differences in multiple comparisons were assessed by one-way analysis of variance. Kaplan–Meier analysis was used to plot survival curves, and log-rank test was used for analysis. Comparisons among categorical variables were performed using χ² or Fisher’s exact tests. A two-tailed P-value of 0.05 or less was considered statistically significant.

Reanalysis of the GEO datasets (GSE21510, GSE22598, GSE23878, GSE32323, and GES39582) showed that DNAJC3-AS1 was significantly upregulated in CRC tissues compared with that in non-tumor tissues ( Figures 1A–E ).

To verify the results of the datasets, we collected 36 CRC tissues paired with normal tissues and detected the expression of DNAJC3-AS1 by qRT-PCR. The expression of DNAJC3-AS1 in CRC tissues was significantly increased compared with that in the corresponding normal tissues ( Figure 1F ). In addition, DNAJC3-AS1 was significantly upregulated in the four CRC cell lines HT29, HCT116, SW480, and SW620 compared with that in the normal colon cell line CCD-18Co ( Figure 1G ).

We further verified the expression of DNAJC3-AS1 in CRC tissues and its correlation with clinicpathological parameters of patients with CRC using ISH. DNAJC3-AS1 was highly expressed in the cancer nests of colorectal cancer tissues compared with that in the adjacent normal colorectal tissues ( Figure 2A ). We analyzed the relationship between DNAJC3-AS1 expression and clinicopathological parameters, such as the level of local invasion (T stage), lymphatic invasion (N stage), and BMI. The data showed that DNAJC3-AS1 expression was positively correlated with local invasion (T stage), I–IV stage, and BMI ( Figures 2B, C , Supplementary Table S1 ). High expression of DNAJC3-AS1 was also associated with shortened overall survival of patients with CRC, suggesting that upregulation of DNAJC3-AS1 is a predictor of poor prognosis ( Figure 2D ). The above results indicated that the high expression of DNAJC3-AS1 in CRC is associated with higher local infiltration, TNM stage, BMI, and poor prognosis.

To further analyze the biological function of DNAJC3-AS1 in CRC, HCT116 and SW480 cells were transfected with siRNA to knock down the expression of DNAC3 AS1 in CRC cells. The transfection effect was detected by qRT-PCR. The results showed that after siRNA transfection, the expression of DNAJC3-AS1 in HCT116 and SW480 cells was significantly inhibited ( Figure 3A ). CCK-8 assays revealed that knockdown of DNAJC3-AS1 significantly inhibited the proliferation of HCT116 and SW480 cells ( Figure 3B ). Results of the colony formation assay were consistent with those of the CCK-8 assay; the knockdown of DNAJC3-AS1 significantly reduced the number of clone colonies ( Figure 3C ).

We performed a wound-healing assay to analyze the influence of DNAJC3-AS1 on the migration ability of CRC cells. The results showed that knockdown of DNAJC3-AS1 can significantly reduce the migration ability of HCT116 and SW480 cells ( Figure 4A ). Finally, we used the Transwell assays to analyze the effect of DNAJC3-AS1 on the invasion ability of CRC cells. The results showed that knockdown of DNAJC3-AS1 significantly inhibited the invasion ability of HCT116 and SW480 cells ( Figure 4B ). Therefore, the above results confirmed that DNAJC3-AS1 knockdown was effective in CRC cells and could inhibit the migration and invasion abilities of CRC cells. Therefore, the high expression of DNAJC3-AS1 serves as an oncogene in CRC.

Based on our previous finding that the higher DNAJC3-AS1expression level correlated with BMI (P=0.038) ( Supplementary Table S1 ). We further evaluate the effects of DNAJC3-AS1 on lipid content in CRC cell lines SW480. Our data demonstrated that DNAJC3-AS1 knockdown induced decreased levels of lipid accumulation (Nile red staining) in SW480 ( Figure 4C ).

To provide further evidence, we measured the expression of FFA metabolic enzymes and related signaling pathway. In our study, western blotting showed that knockdown of DNAJC3-AS1 significantly reduced the expression levels of P-EGFR, P-PI3K, P-Akt, P-NF-κB, and SREBP1 in HCT116 and SW480 cells ( Figure 5 ). The EGFR/PI3K/AKT/sterol regulatory element-binding protein (SREBP1) signaling pathway is closely related to lipid metabolism. Through western blotting, we found that knockdown of the DNAJC3-AS1 gene decreased the expression levels of ACC1 and FASN in HCT116 and SW480 cells ( Figure 5 ). Therefore, DNAJC3-AS1 may regulate the expression of ACC1/FASN via the EGFR/PI3K/AKT/NF-Kb/SREBP1 pathway to promote the progression of CRC.