Association of SEPT9 Gene Methylation with the Clinicopathologic Features and Fusobacterium nucleatum Infection in Colorectal Cancer Patients

Abstract

Background/Objectives: Colorectal cancer (CRC) is a significant global health issue. The identification of methylated Septin 9 (mSEPT9) as a biomarker for CRC represents a significant advancement in cancer diagnostics. On the other hand, Fusobacterium nucleatum (FN) is one of the most studied cancer-related microbes in CRC. This study provided cohort evidence on the association of mSEPT9 with clinicopathologic characteristics and FN infection in CRC patients. Methods: Paired formalin-fixed paraffin-embedded (FFPE) tissue DNA (cancerous and adjacent non-cancer tissues) of eighty-three CRC patients was collected. Methylation-specific qPCR targeting the v2 promoter region of mSEPT9 was carried out on bisulfite-converted FFPE DNA. For FN detection, a TaqMan probe-based method targeting the 16S rRNA gene was used. The differences in mSEPT9 levels and FN expression between cancer and non-cancer tissues were evaluated. Association studies between mSEPT9 in the tumor and relative mSEPT9 levels with FN infection and available clinical data were conducted. Results: Higher mSEPT9 levels were found in the cancerous tissue compared to non-cancerous tissue (p < 0.0001). High mSEPT9 levels in the tumor were significantly associated with older patients (p < 0.001) and larger tumor size (p = 0.048) but not with other clinicopathologic variables. In double-positive patients where mSEPT9 was detected in both cancerous and non-cancerous tissue, the expression fold-change in mSEPT9, calculated using the 2^−ΔΔCT formula, was significantly higher in patients with tumor size equal to or greater than 5 cm (p = 0.042). High levels of mSEPT9 in tumor were not associated with FN infection. However, high levels of FN infection were associated with mSEPT9 (p < 0.021). Conclusions: High levels of mSEPT9 are found in CRC tumor tissue and are associated with older age and larger tumor size, while high levels of FN infection are associated with mSEPT9 in this single-center cohort study.

1. Introduction

Colorectal cancer (CRC) is among the most common cancers globally, ranking as the third most frequently diagnosed cancer in both sexes combined, and it caused the second highest cancer-related deaths after lung cancer in 2022 [1]. In Malaysia, it is the second most common cancer after breast cancer, according to the most recent National Cancer Registry Report [2]. The standard detection markers for CRC patients using their blood are carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA 19-9) [3], while the less invasive stool-based biomarkers are occult blood detection using the fecal immunochemical test (FIT) method and commercially available methylated Vimentin (VIM) detection kit [4]. Similarly, detection of methylated Septin 9 (mSEPT9) has also been FDA-approved as a blood-based biomarker for CRC screening [5,6]. SEPT9 is a tumor suppressor gene that belongs to the SEPTIN family, which regulates various cellular functions such as DNA repair, cytokinesis, chromosome segregation, and apoptosis [7,8]. The dysregulation of SEPT9 expression through a high level of methylation (hypermethylation) can lead to disruptions in their tumor-suppressing functions, thus contributing to cancer progression [8]. This forms the basis of hypermethylated SEPT detection in the blood of CRC patients.

Fusobacterium nucleatum (FN) is a Gram-negative anaerobic bacterium that is ubiquitously found in the human oral cavity. Abundant growth of FN can contribute to inflammatory diseases such as periodontitis and gingivitis [9,10,11]. Extensive studies have demonstrated that FN is a common gut microbiome, and the bacterial infection plays a role in the initiation and progression of CRC by modulating immune cells [12] and activating inflammation [13], to collectively establish a tumor-promoting microenvironment [14,15]. Our group has recently reported the high FN load in CRC tissues and its association with lymph node metastasis and cancer staging in a small cohort study [16]. In a previous study, Tahara et al. demonstrated that FN could accelerate DNA methylation of specific genes in the colonic mucosa in ulcerative colitis [17]. In CRC, FN was found to associate with CIMP, which is the hypermethylation of CpG islands in the promoters of tumor suppressor genes such as MYOD1, MGMT, N33, and so on [17]. High FN was also associated with methylation of CDKN2A [18] and MLH1 [17,19] in CRC. So far, no studies have evaluated the association between FN infection and SEPT9 methylation despite its crucial clinical value. In this study, we demonstrated the high level of SEPT9 methylation in the CRC tumor tissue compared with the adjacent non-cancer tissue in our cohort study. We also reported the association of high FN load with SEPT9 methylation.

2. Materials and Methods

2.1. FFPE Tissue Collection, Staining and DNA Extraction

FFPE tissue blocks from a small cohort consisting of a total of 83 CRC patients, along with the corresponding clinicopathologic information, were collected from a single medical center as described previously [16]. Hematoxylin and eosin (H&E) staining was performed on the tissue slides following a previous protocol [20,21], and both cancerous and adjacent non-cancer tissues were independently reviewed and confirmed by two senior pathologists. Adjacent tissue is defined as healthy, non-cancerous colonic cells (confirmed by H&E staining) that are found adjacent to the cancerous colon tumor cells. Total FFPE tissue genomic DNA (gDNA) was extracted using the ReliaPrep FFPE gDNA MiniPrep System (Promega, Fitchburg, WI, USA), and the DNA yield and quality were measured using the SpectraMax QuickDrop spectrophotometer (Molecular Devices, San Jose, CA, USA).

2.2. Detection of Methylated SEPT9 by qPCR

Extracted FFPE DNA was bisulfite-converted using the EZ DNA Methylation-Lightning Kit (Zymo Research, Irvine, CA, USA). Methylation-specific qPCR was used to detect the v2 promoter region of mSEPT9 together with the bisulfite-converted human endogenous control, ACTB. 1 µL of bisulfite-converted DNA was assayed in 20 µL of Environmental Master Mix (Applied Biosystems, Waltham, MA, USA) using the QuantStudio3 (Thermo Fisher Scientific, Waltham, MA, USA) with the following reaction conditions: mSEPT9; 10 min activation at 95 °C, 40 cycles of 15 s at 95 °C, and 45 s at 50 °C; and bisulfite-converted ACTB– 10 min activation at 95 °C, 40 cycles of 15 s at 95 °C, and 45 s at 56 °C. The results were validated with no-template control, positive methylation control (ATCC HCT116 gDNA), and negative methylation control (ATCC CCD-112 gDNA). All reactions were run in triplicates, and mSEPT9 detection is considered valid if the endogenous control has a CT value below 40. The relative methylation levels of SEPT9 in each sample were expressed as ΔCT in reference to bisulfite-converted ACTB, while the expression fold-change in mSEPT9 between cancerous and non-cancerous tissue was represented by 2^−ΔΔCT. Primers and probes were purchased from Integrated DNA Technologies (IDT). The sequences are presented below:

mSEPT9 v2 promoter forward primer: 5′ AAATAATCCCATCCAACTA 3′, reverse primer: 5′ GATT-DS-GTTGTTTATTAGTTATTATGT 3′, probe: 5′ FAM-TTAACCGCGAAATCCGAC-TAMRA 3′, blocker: 5′ GTTATTATGTTGGATTTTGTGGTTAATGTGTAG-C3 3′.

Bisulfite-converted ACTB forward primer: 5′ TGGTGATGGAGGAGGTTTAGT AAGT 3′, reverse primer: 5′ AACCAATAA AACCTACTCCTC CCT TAA 3′, probe: 5′ FAM-ACCACCACCCAACACACA ATAACAAACACA-TAMRA 3′.

2.3. Detection of Fusobacterium nucleatum (FN) by qPCR

FN was detected using Taqman probe-based qPCR as described in the previous protocol [16]. Briefly, the 16S rRNA gene of FN was measured along with the human endogenous control, the 18S rRNA gene. A total of 80 ng FFPE DNA template was assayed in 20 μL of Environmental Master Mix 2.0 (Applied Biosystems, USA) using the QuantStudio3 (ThermoFisher Scientific, USA) with the following reaction conditions: 10 min activation at 95 °C, 40 cycles of 15 s at 95 °C, 30 s at 48 °C, and 1 min at 60 °C. The relative abundance of FN in each sample was expressed as ΔCT in reference to 18S rRNA, while the expression fold-change in FN between cancerous and non-cancerous tissue was represented by 2^−ΔΔCT. Primers and probes were purchased from Integrated DNA Technologies (IDT). The sequences are presented below:

FN 16S rRNA forward primer: CCATTACTTTAACTCTACCATG, reverse primer: CTGAGGGAGATTATGTAAAAATC, probe: FAM-CAATTTCAG-ZEN-CAACTTGTCCTTCTTGATC-3IABkFQ.

18S rRNA forward primer: GAGACTCTGGCATGCTAACTAG, reverse primer: GGACATCTAAGGGCATCACAG, probe: Cy3-TGCTCAATCTCGGGTGGCTGAA-3IAbRQSp.

2.4. Association Studies and Statistical Analysis

All statistical analysis was performed as described previously [16] using Prism GraphPad (8.0) for Windows. Categorical data were analyzed using the chi-square test (two-tailed). Continuous data were analyzed using either the Mann–Whitney (two-tailed) or Wilcoxon signed-rank test (two-tailed). A p-value of <0.05 was considered statistically significant.

3. Results

3.1. Detection of mSEPT9 in CRC and Their Association with Clinicopathologic Features

Clinicopathologic variables of CRC patients in this cohort study are summarized in

Table 1. Association of methylated SEPT9 with clinicopathologic features.

Feature	mSEPT9 Detection, n (%) 1		Total, n (%) 1	p Value 2
	Present (n = 75)	Absent (n = 8)
Gender
Male	32 (38.55)	5 (6.02)	37 (44.57)	0.283
Female	43 (51.81)	3 (3.61)	46 (55.43)
Age
<50	9 (10.84)	1 (1.20)	10 (12.05)	0.967
≥50	66 (79.52)	7 (8.43)	73 (87.95)
Tumor site
Colon	53 (64.63)	4 (4.88)	57 (68.67)	0.207
Rectum	21 (25.61)	4 (4.88)	25 (30.12)
Unknown	1	0	1 (1.21)
Tumor size
<5 cm	43 (54.43)	7 (8.86)	50 (60.24)	0.134
≥5 cm	28 (35.44)	1 (1.27)	29 (34.94)
Unknown	4	0	4 (4.82)
Cancer staging
I + II	18 (21.69)	4 (4.82)	22 (26.51)	0.113
III + IV	57 (68.67)	4 (4.82)	61 (73.49)
KRASstatus
Wildtype	38 (46.91)	5 (6.17)	43 (51.81)	0.574
Mutated	35 (43.21)	3 (3.70)	38 (45.78)
Unknown	2	0	2 (2.41)
pT
T2 + T3	55 (66.27)	5 (6.02)	60 (72.29)	0.515
T4	20 (24.10)	3 (3.61)	23 (27.71)
pN
N0	19 (22.89)	4 (4.82)	23 (27.71)	0.138
N1 + N2	56 (67.47)	4 (4.82)	60 (72.29)
pM
MX	65 (78.31)	8 (9.64)	73 (87.95)	0.271
M1	10 (12.05)	0 (0.00)	10 (12.05)
FN
Detected	67 (80.72)	6 (7.23)	73 (87.95)	0.237
Not detected	8 (9.64)	2 (2.41)	10 (12.05)

¹ n represents number of cases. (%) represents percentage. ² Chi-square test (two-tailed) of methylated SEPT9 with clinicopathologic features.

. Notably, mSEPT9 was mainly detected in CRC tumor tissues in the presence of FN (80.72%), while only 7.23% of FN-positive tissues did not bear SEPT9 methylation. However, the presence of mSEPT9 in the CRC did not show any significant association (p > 0.05) with the clinicopathologic variables, including the FN infection.

3.2. Detection of mSEPT9 in Colorectal Cancer Patients

Out of the 83 patient samples analyzed, 75 (90.36%) tested positive for mSEPT9. Among these positive samples, 30.12% showed detectable mSEPT9 in both cancerous and adjacent non-cancer tissue (referred to as double positive hereafter), while an additional 56.63% exhibited mSEPT9 exclusively in cancerous tissue (Figure 1a). The overall methylation levels of mSEPT9, indicated by ΔCT values, were significantly higher in cancerous tissue compared to non-cancer tissue, as evidenced by the Wilcoxon signed-rank test (p < 0.0001) (Figure 1b). In cases where mSEPT9 was detected in both cancerous and non-cancerous tissue (double positive, n = 25), mSEPT9 was predominantly represented in cancerous tissue (72%) according to the 2^−ΔΔCT method (Figure 1c).

3.3. Differential Expression of mSEPT9 in Tissues

The differences in expression fold-change in mSEPT9 expressed as 2^−ΔΔCT were compared in various clinicopathological variables using the Mann–Whitney test. The fold-change in mSEPT9 was significantly higher in CRC patients aged above or equal to 50 years (p < 0.001) and in those with larger tumors (p = 0.048) (Figure 2a). Among patients who were double-positive for mSEPT9 (n = 25), the fold-change in SEPT9 methylation was significantly greater in those with tumor sizes of 5 cm or larger (p = 0.042). The Mann–Whitney test cannot be performed on FN, age, and distance metastasis groups due to vast differences within the small sample size, hence termed as ‘not available’ (N/A) (Figure 2b).

In our previous work [16], we performed an analysis on the differential expression of FN in CRC tissues without its association with the SEPT9 methylation. Here, we show that a higher expression fold-change in FN in cancer tissues is significantly associated with the SEPT9 methylation (p = 0.021) (Figure 2c).

4. Discussion

In our study of CRC, SEPT9 methylation did not show significant associations with clinicopathologic features. This was contradictory with findings from two studies that reported that SEPT9 methylation correlated with larger tumor size, advanced cancer staging including TNM stages, distant metastasis, and specific gene mutations such as TP53, BRAF, PIK3CA, and MMR status [22,23]. These associations suggest a role for SEPT9 methylation in promoting CRC aggressiveness and progression. The differences in findings between our study and others may stem from factors such as varying sample sizes and study populations. Our study may not have had sufficient sample size to support the association studies between SEPT9 methylation and clinicopathologic features. Additionally, genetic and environmental differences among study populations could influence these outcomes.

Our study showed that higher mSEPT9 levels in CRC tissue are associated with older age, especially those over 50. This supports CRC screening recommendations using mSEPT9 detection for individuals aged 50 and above [24,25]. High levels of mSEPT9 are linked to larger tumor size in CRC, reflecting SEPT9′s pivotal role in cytokinesis, the final stage of cell division where the cytoplasm divides to form two daughter cells. SEPT9 orchestrates the assembly and organization of the contractile ring critical for cytokinesis. Dysfunctional SEPT9 can disrupt this process, leading to incomplete cell division and unstable cellular conditions. This disruption may lead to multinucleated cells and abnormal structures observed in cancerous tissues [26], potentially contributing to larger tumor masses in CRC. SEPT9 is also important in autophagy, an essential process for cellular stability. It interacts with autophagy-related proteins, affecting cellular component recycling. Dysfunctional autophagy is linked to cancer development by causing cellular stress and genomic instability. SEPT9′s involvement in these processes suggests it could influence tumor growth in CRC [27,28].

The role of FN in CRC pathogenesis has been previously illustrated [12,13,16]. Additionally, FN could accelerate the methylation of tumor suppressor genes, such as MYOD1, MGMT, N33, CDKN2A, and MLH1, in supporting CRC progression [17,18,19]. Our study is the first to link a high FN bacterial load with SEPT9 methylation in CRC. However, the mechanism behind FN’s influence on SEPT9 methylation is yet to be fully understood and warrants further research.

5. Conclusions

Our findings support mSEPT9 as a biomarker for CRC screening, especially in individuals aged 50 and older. The association of elevated mSEPT9 levels with older age underscores its clinical relevance. Furthermore, SEPT9′s roles in cytokinesis and autophagy suggest its critical impact on cellular stability and cancer pathogenesis. Dysregulation of SEPT9 may contribute to larger tumor sizes and abnormal cellular structures in CRC. Additionally, the novel association of high FN bacterial load with SEPT9 promoter methylation suggests a potential avenue for therapeutic intervention in CRC through targeted antibiotic treatment.