Evaluation of NEUROG1 methylation status in stool specimens in the advanced adenomas and colorectal cancer

Patients and sample selection

The study protocol is shown in Fig. 1, the formalin-fixed, paraffin-embedded (FFPE) samples were collected and stored at room temperature from a total of 75 patients, 25 with CRC, 25 with non-advanced adenomas (NAA), and 25 with advanced adenomas (AA) were diagnosed at the First Affiliated Hospital of Henan University of Science and Technology from May 2019 to May 2020. The patients were confirmed by two experienced doctors, depending on the colonoscopy results and along with the pathology results of CRC, AA, and NAA.

Stool samples were collected from 272 patients diagnosed and treated at Fujian Provincial Hospital from July 2019 to December 2023, including 104 CRC patients, 92 AA patients, and 39 NAA patients. None of the patients had received any anticancer treatment before admission. Control stool samples were collected from 37 healthy individuals undergoing colonoscopy, and CRC, AA, NAA patients were excluded.

In addition, several clinical characteristics were collected, including age, sex, and classification.

This study was approved by the Institutional Review Board of Fujian Provincial Hospital (K2019-11-027). Informed consent was obtained from all enrolled patients and healthy control subjects.

Stool samples were collected before tumor resection and stored at −80 °C in a storage buffer.

DNA extraction from stool

Stool DNA was extracted according to the operation manual and published protocol using a DNA extraction kit in the stool (Ahlquist et al., 2012b). After thawing, the buffered stool samples were homogenized with an oscillator and centrifuged. Then, an aliquot of 12 mL stool supernatant was treated with 50 mg/mL polyvinylpoly pyrrolidone PVPP (Aladdin, Shanghai, China).

The target gene sequence was directly captured by hybridization with the oligonucleotide probe (CGTGCAGCGCCCGGGTATTTGCATAATTTATGCTCGCGGGAGGCCGCCATCGCCCCTCCCCCAACCCGGAGTGTGCCCGTAATTACCG).

For this purpose, 10 mL polyvinylpolypyrrolidone-treated supernatant was denatured using 2.4 M (molar) guanidine isothiocyanate at 92 °C for 15 min (Aladdin, Shanghai, China). Next, 50 µL oligonucleotide capture probe-modified carboxyl magnetic beads (JSR) were added, mixed, and incubated at room temperature for 30 min.

The beads were collected using the magnetic rack and washed with washing buffer (10 mM MOPS, 150 mM NaCl, pH 7.5) three times. Finally, 50 µL nuclease-free water containing 20 ng/µL transfer RNA (Merck K GaA, Darmstadt, Germany) was added to the eluted DNA.

The 50 µL elution was used for bisulfite conversion, and the purified DNA was eluted to 60 µL in the elution buffer. The bisulfite transformation Kit (Zymo Research, Irvine, CA, USA) was used for DNA bisulfite transformation and purification of transformation products. All steps were performed following the manufacturer’s instructions.

DNA extraction of FFPE specimens

TIANamp FFPE DNA Kit (Tiangen Biotech Co., Ltd, Beijing, China) was used to isolate DNA from FFPE tissue samples. Briefly, 4–5 sections of FFPE specimens were collected in 1.5 mL centrifuge tubes. Deparaffinization and dehydration were performed by adding dimethylbenzene and 100% ethanol. After air drying, 400 µL digest buffer with 20 µL proteinase K was added to the precipitation. Then, the precipitation was suspended and digested at 55 °C for one hour. Following a one-hour incubation at 90 °C, the suspension was added into spin columns. After cleaning and centrifugation, the spin columns were dried and eluted with 50–100 µL elution buffer. The DNA samples were stored at −20 °C.

After extraction, the concentration of DNA was determined with the Qubit 2.0 Fluorometer (Thermo Fisher Scientific Inc., California, USA). One µg of extracted DNA was treated with bisulfite using the EZ DNA Methylation-Gold Kit (Zymo Research, Irvine, CA, USA).

Methylation-specific Quantitative PCR (MSP)

The methylation status of NEUROG1 was evaluated by Methylation-specific quantitative PCR (MSP) (Suzhou MicroDiag Biomedicine Co., Ltd, Suzhou, China). Sequences of the primers and probes for NEUROG1 and β-Actin (ACTB) were designed and synthesized by GENEWIZ, (Suzhou, China). The ACTB gene, located on chromosome 7 (7p22.1), was used as a reference. The sequences of the primers and probes for the indicated genes are as follows: NEUROG1_Forward: TCGTGTAGCGTTCGGGT, NEUROG1_Reverse: CACTCCGAATTAAAAAAAAAACG, NEUROG1_probe: ATCCCGCGAACATA; the sequences of predicted promoters and amplified regions in this study are as follows: the purple font is the predicted promoter, and the blue font is the amplified region in this study.

The amplified region is upstream from 172 to 243 of the promoter, i.e., −172 to −243

>hg19_dna range=chr5:134871000-134871843 5′pad=0 3′pad=0 strand=+ repeatMasking=none

CTTGGTGTCGTCGGGGAACGAGGGCAGCACGCTGCGCAGTGCGTCCAGGGCC GCGTTCAAGTTGTGCATGCGGTTGCGCTCGCGATCGTTGGCCTTGACGCGCCGG CTCCTGCGCAGCGAGTGCAGCAGCGCCTCGGAGCGGACCCGCGTCCGGCCG CGGCGCCGCCGCCTCTCCTGCTCGTCGTCCTGTGCCCCTGGAACCTCAGACG CCCGGGAGATATTGGGCGCGCCCCTGCGGGCCGGCGCGGGCGGCCCCGAAGCG GAGGCTGCCTGTTGGAGTCTGGCACAGTCTTCCTCGTCGGTGAGGAAGCCGGAT AGGTCACTGCCGCTGCTGCTGGCGCAGTCGAGGTCGGAGATGCAGGTCTCAAGGC GGGCTGGCATCGTTGCGCTGTGCAGGACCGACGGACAGATAGAAAGGCGCTCAGA GCGCTGCAGCCCGGACTGAGGGCAGAGCCGCCAGGGCGCACTTACGTTCCCAA CAGCCTGGGGTTGTTACTCTGTGCCAGTTGCGGGTGCGAGAGCCTGGAAGGG TGCAGGGGCGCACGGAGAACTTGGCCTGGCCTCCTCGCCTCGCCTGCAGGGG CCACGCGCCCGGCCGGTCTCCTGAGTGATGTCGCCGGCGATCAGATCAGCTCGTG TGAGCACCGAGTGTGGCACACGACTGGCCTCAGGACCCCTTAAGTACCCGGCGC AACAATGGGCGCCCCCCTCCCTTGCCACCTCCGCCCCCGCGGCAGCCCGGGTGA ATGGAGCGAGGCGGCAGGTCATCCCCGTGCAGCGCCCGGGTATTTGCATAATTT ATGCTCGCGGGAGGCCGCCATCGCCCCTCCCCCAACCCGGAGTG ACTB_Forward: GTGATGGAGGAGGTTTAGTAAGTT, ACTB_Reverse: CCAATAAAACCTACTCCTCCCTTAA, ACTB_probe: ACCACCACCCAACACACAATAACAAACACA. In brief, the total volume of qPCR was 30 µL, including 15 µL of DNA sample and 15 µL of PCR master mix. Next, qPCR was conducted on a LightCycler 480 II thermal cycler (Roche Diagnostics Corporation, Indianapolis, USA) using the following conditions: denaturation 95 °C for 20 min one cycle; 45 cycles (annealing 95 °C for 20 s and extension 60 °C for 35 s; and finally chilled to 40 °C for 30 s). After the amplification, the data were analyzed on LC480-II (Roche Diagnostics Corporation, Indianapolis, USA).

A standard curve was elaborated for NEUROG1 methylation to calibrate the investigation system (100%–0% bisulfite-converted NEUROG1-positive genomic DNA; slope = −0.2762; R² = 0.9906). In brief, bisulfite conversion was performed of genomic DNA extracted from cells, in which the NEUROG1 promoter region was fully methylated (HCT116 cells) or unmethylated (293T cells) using a commercially available kit (EZ DNA Methylation-Gold Kit, Enzo, USA). Then, bisulfite-converted fully methylated genomic DNA was mixed with bisulfite-converted unmethylated DNA to obtain a bisDNA gradients (where NEUROG1 methylated DNA was 100%, 50%, 25%, 10%, 5%, 1%, 0.5%, 0.1%, 0.05%, and 0% to the total mixture), while the whole mixture’s bisDNA concentration was maintained at two ng/µL. After that, the samples were subjected to amplification and the results were plotted to generate the standard curve. Negative and positive cell DNA were purchased from Fubio Biotechnology, Co., Ltd, (Suzhou, China).

The quality control’s cycle threshold/crossing point (CP) value is 27–30 for tissue methylation DNA. For stool DNA detection, the CP value of the quality control needs to be <36. For samples without amplification in the target channels, 45 (maximum amplification cycle number of PCR) should be taken for the fitting calculation.

SPSS software (IBM Corp., Armonk, NY, USA) was used to conduct the fitting logistic regression analysis of CP values for target genes and internal reference genes, and the fitting formula was obtained: ${\displaystyle \frac{1}{1+e^{-\left(7.867-0.097\ast \text{NEUROG}1-0.098\ast \text{ACTB}\right)}}.}$

The sample is positive if the fitting value is > 0.7692.

Statistical analysis

Statistical analysis and receiver operating characteristic (ROC) curves were performed using SPSS software version 22.0 (IBM Corp., Armonk, NY, USA). The cut-off value was determined by using the maximum principle of the Youden’s index. The detection performance of stool NEUROG1 gene methylation in CRC and precancerous lesions was investigated. All the data were presented in percentages (%). The Chi-square test was used to compare the data, and P-values <0.05 were considered statistically significant.

NEUROG1 methylation detection in FFPE specimens

To elucidate the methylation status of the promoter region of NEUROG1 in colon cancer, MSP was conducted in both FFPE specimens and patients’ feces samples. We determined the detection rates of NEUROG1 methylation in 75 FFPE samples, including (25 CRC, 25 AA, and 25 NAA) as shown in (Table 1). The methylation rates for CRC was 36%, and for NAA was 24%. Notably, the methylation level of AA was 88% and exhibited positive NEUROG1 methylation. NEUROG1 methylation detection rate was significantly different between CRC+AA group and NAA group in FFPE specimens, as shown in (Table 2).

Stool DNA-NEUROG1 methylation detection in CRC

The methylation level of the NEUROG1gene was evaluated by collecting stool samples from 272 patients, including (104 CRC, 92 AA, 39 NAA) and 37 healthy cohorts. The clinical characteristics of these patients are listed in Table 3. The ages ranged from 18 to 87, with 153 males and 119 females. Next, NEUROG1 methylation in stool samples was tested in different cancer stages.

In the 104 CRC patients, the NEUROG1 methylation level was 63.46%. To further evaluate the sensitivity of NEUROG1 methylation for different tumor classifications and positions, the methylation level was assessed in various CRC stages (I, II, III, IV), as shown in (Table 4).

In the 94 patients with known stages, the detection rate ranged from 41.18% to 70.37%, with no significant differences between each subgroup. The methylation rates of different tumor positions were explored among 100 patients. The methylation rates of proximal colorectal cancer was 48.15%, and for distal colorectal cancer, it was 67.12%. However, the methylation rates in CRC patients with distal CRC seemed higher, and the difference was not statistically significant, as shown in (Table 4).

Stool DNA-NEUROG1 methylation detection in AA and NAA

In 92 cases of AA, the methylation level was 47.83%. Further analysis was conducted to compare the rates in different subtypes of AA based on the tumor position and size. NEUROG1 methylation rates of proximal and distal AA were 36.67% and 57.41%, respectively (P > 0.05). Additionally, when the diameter of the adenoma was ≥3 cm, the detection rate increased significantly, reaching up to 61.90%, as shown in (Table 5). However, there were no significant differences between the different tumor sizes.

In the 39 NAA cases, 27 samples were negative, resulting in a detection specificity of 69.23%. In the 37 healthy volunteers, 31 samples were negative, resulting in a detection specificity of 83.78%. Hence, the NEUROG1 methylation detection sensitivity was 56.12% (110/196) in the positive group of 196 CRC and AA. By comparison, the detection specificity was 76.32% (58/76) in the cases of NAA and healthy people in the control group (Table 6).

Diagnostic performance of NEUROG1 methylation detection with stool samples

Receiver operating characteristic (ROC) curves were elaborated to evaluate the discriminatory capacity of NEUROG1, as shown in (Fig. 2). In our study, 272 stool samples were subjected to NEUROG1 methylation detection. Among those from NAA, healthy individuals were selected as controls and ROC curve analyses were performed. As shown in Fig. 2, the area under the curve (AUC) for the NEUROG1detection of CRC and AA was 0.699 (95% confidence interval (CI) [0.633–0.764]).

In the positive group (CRC and AA), the NEUROG1 methylation test demonstrated a sensitivity of 56.12% (95% CI [49.08%–62.99%]), a positive predictive value (PPV) of 85.94% (95% CI [78.96%–91.47%]), and a positive likelihood ratio (PLR) of 2.37 (Table 6).

In the control group (NAA and healthy volunteers), the test showed a specificity of 76.32% (95% CI [65.45%–85.31%]), a negative predictive value (NPV) of 40.28% (95% CI [32.31%–48.57%]), and a negative likelihood ratio (NLR) of 0.57 (Table 6).

NEUROG1 methylation rate was significantly different between CRC+AA group and NAA+ healthy volunteers group in stool samples, as shown in (Table 7).

Comparison of methylation detection between fecal DNA and serum DNA showed that the detection sensitivity and specificity in fecal DNA was better than in serum DNA.As shown in (Table 8). Therefore, NEUROG1 has the potential to serve as a marker for CRC and AA screening in stool samples.