Molecular and regional characterization of colorectal polyps: insights from proteomics, phosphoproteomics, and immune profiling

Introduction

Background

Colorectal cancer (CRC) remains one of the most common and deadly malignancies worldwide. Overall CRC incidence and mortality have declined among older adults whereas an opposite trend is seen in younger adults. Early-onset CRC exhibits distinct clinical and molecular features compared to later-onset disease; tumors in younger patients behave more aggressively and more likely reflect genetic predisposition (1).

Lifestyle modulators of CRC risk are amplified in individuals with underlying genetic susceptibility but can be modified and thus substantially reduce CRC risk. Mechanistically, lifestyle and environmental factors appear to interface colon tumorigenesis risk through chronic inflammation and microbiome perturbations (2,3).

The predominant pathway leading to CRC is the adenoma-carcinoma sequence; normal colon epithelium progresses to adenomatous polyp and then carcinoma through the acquisition of mutations in tumor suppressors and oncogenes (4). Individuals with familial adenomatous polyposis (FAP), inherit pathogenic variants in the Apc tumor suppressor allele and essentially begin this cascade at a young age leading to CRC in early adulthood. In contrast to this chromosomal instability (CIN) pathway typical of most left-sided tumors, a subset of CRCs especially in the right (proximal) colon express high microsatellite instability (MSI) reflecting a different route of carcinogenesis (5,6). These mechanistic differences by tumor location are mirrored by differences in prognosis and therapy response. Even though the colon is a continuous organ, the right and left segments differ markedly in their biology and the anatomical localization of CRC is thus increasingly identified as clinically relevant (Figure 1).

Figure 1 CRC phenotype differences by anatomic location (created using BioRender). CIN, chromosomal instability; CRC, colorectal cancer; IMA, inferior mesenteric artery; MSI, microsatellite instability; SMA, superior mesenteric artery.

These regional differences translate into distinct local microenvironments, both microbiological and immunological, that can modulate tumor development (7). Right- versus left-sided CRCs may thus follow separate evolutionary routes, underscoring the relationship with epidemiologic trends and outcome disparities (8). Recognizing these intrinsic regional differences is crucial for interpreting CRC pathogenesis and tailoring treatment strategies.

Disrupted cell signaling networks, specifically those regulated by protein kinases are a hallmark of colorectal tumorigenesis. Aberrant kinase activity drives uncontrolled proliferation, survival, and genomic instability in evolving tumors (9). Conversely, loss or inhibition of DNA damage checkpoint kinases such as ATR that safeguard genome integrity by activating DNA damage checkpoints impairs DNA repair and can accelerate the accumulation of mutations (10). We hypothesized that important growth and stress-response kinases such as CDK4 and ATR might be differentially engaged in polyps versus normal mucosa, and potentially between regions of the colon. This is reflected by observations on large cancer cohorts demonstrating that CDK4 and ATR expression levels are often elevated in colorectal tumors compared to normal tissue. Because these kinases represent attractive therapeutic targets, defining whether oncogenic signals predominate in right- versus left-sided polyps offers avenues for precision prevention and treatment strategies (11,12). DNA damage accumulation, instigated by unrepaired DNA damage is a defining feature of colorectal neoplasia. Failure to repair DNA double-strand breaks (DSBs) leads to chromosomal aberrations or cell death and cells express multiple DNA repair pathways (13). Failure in these systems accelerates carcinogenesis. An established molecular marker of DNA damage in tissues is phosphorylated form of histone H2AX. Elevated γH2AX uptake thus indicates ongoing or unresolved DSBs and identifies cells under genomic stress. Environmental factors can influence both the rate of DNA damage and how it is repaired. For example, chronic inflammation or exposure to genotoxic bacterial toxins can overwhelm high-fidelity repair pathways, and force cells into error-prone repair (environmental gear selection) (14,15). Thus, we hypothesized that differences in the colonic milieu from right to left, such as variations in resident bacteria or inflammatory response, might result in different levels of DNA damage and checkpoint activation in polyps. This connection between microbiome, DNA damage, and signaling pathways (such as ATR activation in response to microbiota-induced DNA breaks) is an area of active research and directly ties into our study’s focus.

The colonic microbiome has emerged as a pivotal player in colorectal neoplasia progression. Bacterial metabolites or virulence factors can directly damage DNA or promote pro-carcinogenic inflammation (16). Thus, perturbations in microbial community structure have been linked to the development of adenomas in both sporadic CRC and hereditary polyposis conditions. This insight has raised interest in microbiome-targeted interventions as potential tools for CRC prevention. The impact of dysbiosis and therapeutic measures are therefore region-specific. For example, Fusobacterium nucleatum (Fn), an oral anaerobe is frequently found enriched within colon tumors especially right-sided cancers and in relation with the polymicrobial biofilm juxtaposed on the mucosa (17). Fn can activate oncogenic signaling (such as β-catenin pathways) (18), modulate the local inflammatory milieu, suppress anti-tumor immune responses, and even confer chemoresistance (19). Notably, Fn has been shown to directly induce DNA damage and genomic instability in APC-mutant mouse models, effectively accelerating intestinal tumorigenesis. This bacterium’s dual ability to drive proliferation (through Wnt/β-catenin activation) and inflict DNA damage provides a mechanistic link between dysbiosis, and the molecular changes observed in right-sided polyps (20,21). The predominance of biofilms and organisms like Fusobacterium in the proximal colon may thus help explain why we observe more extensive DNA damage (e.g., higher γH2AX levels) and certain signaling aberrations in right-sided lesions compared to left-sided ones. In tandem with microbes, the immune microenvironment of the colon shows regional variation that can influence tumor behavior.

The immune landscape of the colon also varies by region and may influence tumor initiation. CD103 positive tissue-resident memory T cells (TRMs), regulate immune surveillance and have been associated with better outcomes in CRC (22,23). MSI-high tumors (frequently right-sided) typically contain more TRMs and other tumor-infiltrating lymphocytes (24), reflecting a more immunogenic microenvironment.

Lastly, the proliferative potential of polyps may differ across the colon due to regional variation in stem cell dynamics. Intestinal stem cells (ISCs), marked by LGR5 and other genes, maintain crypt homeostasis but with disrupted Wnt signaling can become tumor-initiating (25,26). FAP polyps, by virtue of APC loss, already exhibit excessive Wnt activation. However, the right colon may provide a more permissive niche for ISC expansion or faster regeneration, possibly explaining our observed differences in polyp-derived organoid growth.

Rationale and knowledge gap

Given the changing epidemiologic landscape of CRC especially the increase in young-onset, left colon predilecting disease our focus on proteomic and phosphoproteomic profiles in polyp vs. non-polyp mucosa with regional differences aims to elucidate whether certain molecular or immune alterations are region-specific and, in turn, whether these changes are driven by dysbiosis. The integration of these concepts with stem cell dynamics and cell growth and repair mechanisms could herald tailored surveillance strategies and treatments that consider both the molecular profile of polyps and the regional context.

Objective

To characterize colonic regional molecular, microbial, immune, and stem cell phenotypic differences between pediatric FAP adenomas and adjacent mucosa using integrated proteomic/phosphoproteomic and functional assays, and to evaluate ATR/CDK4 as candidate chemopreventive targets. We present this article in accordance with the MDAR and ARRIVE reporting checklists (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-138/rc).

Methods

Cohort selection

Pediatric patients aged <21 years with a diagnosis of FAP based on standard clinical criteria (27), followed at the Pediatric Hereditary Polyposis Clinic at Children’s Mercy Kansas City and undergoing (I) colectomy or (II) surveillance colonoscopy as per standard of care were included. Patients were excluded from the study if (I) on (off-label) chemopreventive agents, or recently (1 month) treated with probiotics or antibiotics, (II) diagnosed with attenuated FAP (AFAP), or (III) unrelated colonic disease (e.g., colitis; inflammatory bowel disease). Participation in the study was entirely voluntary, informed consent was obtained from the patient or parent/legal guardian, and assent was obtained from the child, if minor, prior to participation in the study. Given the small number of patients diagnosed with FAP, the sample size was determined by the number of patients available at our center between 2021 and 2024. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics board of Children’s Mercy Hospital (IRB Nos. 15020066 and 00001861) and informed consent was obtained from all individual participants or their legal guardians.

Colectomy derived polyps and adjacent (non-polyp) tissue harvesting

The colon was transected longitudinally and immobilized for tissue sampling. Different segments of the colon were identified by gross morphologic landmarks moving distally from the ileocecal valve. The colonic mucosa was examined by low magnification to determine the presence and size of polyps. Representative samples of normal, i.e., non-grossly polypoid mucosa was obtained from each colonic region identified. Polyps were sampled from different regions—polyps were either shaved off the mucosa or, if pedunculated, resected without pedicle. Polyps (>10 mm in diameter) were transected and submitted for study procedures (proteomics, immunohistochemistry-histology) and for routine histology. Normal-appearing, non-polyp involved mucosa was obtained from different regions and placed into sterile polypropylene tubes, flash frozen in liquid nitrogen, and stored in −80 ℃ freezers until transport. We furthermore created up to five unstained slides of normal mucosa from tissue blocks from each segment of the subject’s colon (cecum, ascending, transverse, descending, sigmoid colon, and rectum—if excised). Our subjects included three patients: 2F, with FAP and Apc PV and aged 15–16 years of age at the time of colectomy.

Endoscopic procedure, tissue harvesting and immediate processing

Polyps (>5–10 mm) in size were removed by standard cold snare polypectomy method described elsewhere (28) and retrieved by suction—polyp trap (Steris^®) method immediately following resection. Cold forceps biopsy of normal (non-polyp) mucosa from the same anatomic region and from within 10 centimeters of the harvested polyp was also obtained. This fresh tissue was collected for organoid experiments and preserved in phosphate-buffered saline (PBS) or Dulbecco’s modified Eagle medium (DMEM) media and delivered on wet ice within 2 h of collection.

Mouse model

Apc^Min/+ mice (Strain #: 002020) were purchased from Jackson Laboratory (Bar Harbor, USA). Mice were maintained in helicobacter and parvovirus-free environment and used between six and eight weeks of age. Wild type mice of identical genetic background were used as control groups. Animal experiments were performed under a project license (No. 23-10-353) granted by University of Kansas Medical Center in compliance with the National Institutes of Health guidelines for the care and use of animals.

Collection of conditioned media (CM) from Fn and western blot analysis

We generated CM by growing Fn anaerobically followed by centrifugation and 0.22 mm filtration. HCT116 colon cancer cells (Cat#: CCL-247, Manassas, USA) were incubated with FnCM (1:1,000 dilution; 10.476 mg/mL protein) for 48 h. Cells were harvested for protein extraction using radioimmunoprecipitation assay (RIPA) buffer. Protein concentration was quantified, after which 50 µg of protein per sample was denatured by heating in sodium dodecyl sulfate (SDS) sample buffer containing β-mercaptoethanol. Proteins were separated according to molecular weight by SDS-polyacrylamide gel electrophoresis and transferred onto a membrane. Following transfer, membranes were blocked to minimize nonspecific interactions and then incubated with primary antibodies directed against ATR (1:500, #2790, Cell Signaling Technology, Danvers, USA), CDK4 (1:500, #12790, CST), and β-actin (1:10,000, A3854). Detection was performed using species-appropriate secondary antibodies, including goat anti-mouse IgG (Cat#: ab6789, Abcam, Cambridge, UK) or anti-rabbit IgG (Cat#: ab7090, Abcam).

Immunostaining

HCT116 cells were exposed to PBS (control) or FnCM for 48 h and cells were stained with antibody to γH2AX (Cell Signaling, 9718, 1:250). For immunofluorescence staining of the polyp or non-polyp tissues, paraffin-embedded sections were deparaffinized, rehydrated, and subjected to antigen retrieval using sodium citrate buffer (pH 6.0) at high temperature. Sections were permeabilized with 0.2% Triton X-100 in PBS for 10 min and tissues were incubated overnight at 4 ℃ with primary antibodies targeting lipopolysaccharide (LPS) (PA1-73178), lipoteichoic acid (LTA) (MA1-7402) and 8-Oxoguanine (MAB3560). After washing, appropriate Alexa Fluor-conjugated secondary antibodies (Cat#: SA000011; Thermo Fisher, Waltham, MA, USA, 1:500) were applied for 1 h at room temperature in the dark. Nuclei were counterstained with DAPI (0.5 µg/mL) for 5 min. Slides were mounted using ProLong Gold Antifade Mount (Thermo Fisher Scientific) and imaged using a Nikon fluorescence microscope (Melville, USA).

For immunohistochemistry of paraffin sections from control or FAP patients or Apc^Min/+ mice (6 to 8 weeks old; Strain #: 002020; Jackson Laboratory), tissues were sectioned at 5 µm thickness and mounted onto Superfrost Plus slides (Thermo Fisher Scientific). For histological assessment, sections were deparaffinized in xylene, rehydrated through a graded ethanol series, and stained with hematoxylin and eosin (H&E) following standard protocols (29). For immunohistochemistry, deparaffinized sections were subjected to heat-mediated antigen retrieval in 10 mM sodium citrate buffer (pH 6.0) using a boiler. Endogenous peroxidase activity was blocked using 3% hydrogen peroxide in methanol for 10 min. Following incubation of sections overnight with primary antibodies at 4 ℃, sections were washed three times for 5 min each in PBS and incubated with biotinylated secondary antibodies for 30 min. This was followed by incubation with enzyme streptavidin-peroxidase conjugate for 30 min. After staining, the sections were mounted with mounting medium, and analyzed with a Nikon 80 microscope.

Flow cytometry

We performed the flow cytometry to isolate immune cells in either control mucosa or polyps. For gating, we utilized fresh biopsies to establish viability before staining cells with antibodies to detect CD4⁺ or CD8⁺ cells (Cat # MA5-17009; Thermo Fisher Scientific) and separated CD103⁺ cells as a subset of CD4 cells.

Organoid assay

Following colonoscopy of a 10-year-old male patient, crypts were isolated from the polyps, embedded in Matrigel (Corning, Corning, USA), and cultured in a stem cell medium (50% L-WRN CM + Y27632 + SB431542) as described (30).

Quantitative real-time polymerase chain reaction (qRT-PCR)

Total RNA was isolated from organoids using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Waltham, USA) and reverse-transcribed into complementary DNA (cDNA) with a High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). qRT-PCR was performed to quantify mRNA expression of target genes using gene-specific primers (Table 1), with GAPDH serving as the internal reference gene. Reactions were prepared using a SYBR Green-based master mix (Thermo Fisher) containing cDNA template and primers and were run on a real-time PCR system (Thermo Fisher). Relative transcript abundance was calculated from threshold cycle (Ct) values using the comparative Ct (2^−ΔΔCt) method.

Proteome and phosphoproteome analysis via the tandem mass tag (TMT) system

Total protein from each sample underwent reduction and alkylation, followed by purification using chloroform/methanol extraction and digestion with MS-grade porcine trypsin/LysC (Promega, Madison, WI, USA). Peptides were labeled with a TMT 10-plex isobaric reagent set (Thermo Fisher Scientific) and sequentially enriched using High-Select TiO₂ and Fe-NTA phosphopeptide enrichment kits (Thermo Fisher Scientific) according to the manufacturer’s protocols. Labeled phosphopeptide-enriched and non-enriched fractions were initially separated into 46 fractions on a 100 mm × 1.0 mm Acquity BEH C18 column (Waters, Milford, MA, USA) using an UltiMate 3000 UHPLC system (Thermo Fisher Scientific) with a 40-min basic-pH gradient from 99:1 to 60:40 buffer A:B, after which fractions were consolidated into 18 super-fractions.

Each super-fraction was further resolved by reverse-phase chromatography on XSelect CSH C18 2.5 µm resin (Waters) using an in-line 75 µm internal diameter column coupled to an UltiMate 3000 RSLCnano system (Thermo Fisher Scientific). Peptides were eluted over a 75-min gradient from 98:2 to 60:40 buffer A:B and introduced by electrospray ionization at 2.4 kV into an Orbitrap Eclipse Tribrid mass spectrometer (Thermo Fisher Scientific) operated with multi-notch MS3 acquisition. Full MS scans were collected using the fourier transform mass spectrometry (FTMS) analyzer in top-speed profile mode at a resolution of 120,000 across an m/z range of 375–1,500. Following collision-induced dissociation (CID) activation at a normalized collision energy of 31.0, MS/MS spectra were acquired in centroid mode using the ion trap analyzer. Synchronous precursor selection was applied to select up to 10 MS/MS precursors for higher-energy collisional dissociation (HCD) activation at a normalized collision energy of 55.0, with subsequent acquisition of MS3 reporter ions using the FTMS analyzer in profile mode at a resolution of 50,000 over an m/z range of 100–500.

Buffer A consisted of 0.1% formic acid and 0.5% acetonitrile, and buffer B consisted of 0.1% formic acid and 99.9% acetonitrile. For offline fractionation, both buffers were adjusted to pH 10 using ammonium hydroxide.

Proteomic and phosphoproteomic data analysis

Protein identification and reporter ion quantification were performed by searching the UniProtKB database restricted to Homo sapiens using MaxQuant (Max Planck Institute, version 2.1.4.0) (31). Database searches employed a parent ion mass tolerance of 3 ppm, fragment ion tolerance of 0.5 Da, and reporter ion tolerance of 0.001 Da, with trypsin/P specified as the proteolytic enzyme allowing up to two missed cleavages. Variable modifications included methionine oxidation, protein N-terminal acetylation, and phosphorylation on serine, threonine, and tyrosine residues, while carbamidomethylation of cysteine was defined as a fixed modification. Protein identifications were accepted at a false discovery rate below 1.0%, and entries identified exclusively by modified peptides were excluded. Protein probabilities were assigned using the Protein Prophet algorithm (32). TMT MS3 reporter ion intensity values from unenriched lysate samples were used to evaluate changes in total protein abundance, whereas phospho-STY modifications were identified from samples enriched for phosphorylated peptides. The enriched and un-enriched samples were multiplexed using two TMT10-plex batches, one for the enriched and one for the un-enriched samples.

After data acquisition and database searching, MS3 reporter ion intensities were normalized using ProteiNorm (33). Data were subsequently normalized with VSN (34) and analyzed using proteoDA (35) to perform statistical testing based on Linear Models for Microarray Data (36), incorporating empirical Bayes (eBayes) smoothing of standard errors. Differential analysis of phosphopeptides followed a comparable framework with additional processing steps. Phosphosites were filtered to retain peptides with localization probabilities greater than 75%, peptides with zero values were removed, and data were log₂ transformed. Limma was additionally applied for differential testing. Proteins and phosphopeptides meeting a false discovery rate (FDR)-adjusted P value <0.05 and an absolute fold change >2 were considered statistically significant. Protein pathway analysis was conducted using Metascape (37) and phosphoproteomic data was further analyzed using KEA3 (38). Proteomic and phosphoproteomic data are available in tables available at https://cdn.amegroups.cn/static/public/10.21037tgh-2025-138-1.xlsx, and via ProteomXchange and the PRIDE database (39-41) via the dataset identifier PXD061190 and 10.6019/PXD061190.

Statistical analysis

Data are presented as mean ± standard deviation (SD) or, for histologic scoring, as standard error of the mean (SEM). Comparisons between two groups were evaluated using an unpaired, two-tailed Student’s t-test. When more than two groups were compared, one-way analysis of variance (ANOVA) was applied, followed by Fisher’s protected least significant difference (PLSD) post hoc testing. A P value <0.05 was considered statistically significant. All statistical analyses were conducted using GraphPad, version 9.

Results

Proteomics and phosphoproteomics of polyp and non-polyp mucosa

We harvested six polyps (mean 14 mm diameter) and 4 normal, non-polyp mucosal specimens from throughout the colon in three patients with FAP and carried out quantitative proteomic/phosphoproteomic studies. Proteins from each of these tissue samples were isolated, digested into peptides and each of the 10 samples labeled with a distinct TMT 10-plex reagent (42). After labeling, samples were combined into a single sample and phospho-peptides were enriched using a sequential TiO₂ and FeNTA enrichment protocol. Next, both the phosphopeptide enriched and the un-enriched samples were subjected to high-pH peptide fractionation after which fractions were analyzed on an Orbitrap Eclipse mass spectrometer. After database researching, quality control and normalization, a total of 5,254 proteins and 4,233 phosphopeptides were detected, identified, and quantified. Figure S1 represents normalized intensities, cluster dendrogram and heatmaps, respectively.

There were 405 proteins whose expression was statistically significantly increased in mucosa over polyps and as shown in Figure 2A,2B. Of these proteins, 239 were more than 2-fold overexpressed in mucosa over polyps. The four most overexpressed proteins in this list were alpha-internexin, chromogranin-A, xylosyltransferase 2, and a fragment of palmitoyl-protein hydrolase. Next, there were also 405 proteins whose expression was statistically significantly increased in polyps over mucosa and as shown in Figure 2B, 149 of these proteins were more than 2-fold overexpressed in polyps over mucosa. The four most overexpressed proteins in this list were olfactomedin-4, histone H2B type 2-E, lipocalin-15, and a fragment of the mitochondrial 39S ribosomal protein L1.

Figure 2 Quantitative proteomics in the polyp vs. the synchronous mucosa. (A,B) Scatter and Volcano plots of adjusted P value vs. fold change of proteins upregulated in polyps (red) or upregulated in mucosa (blue). (C) Computational analysis system Metascape was used to analyze the overexpressed protein pathways in mucosa and polyps. (D) Volcano plot of adjusted P value vs. fold change of phosphopeptides upregulated in polyps (red) or upregulated in mucosa (blue). (E,F) MeanRank score of top 20 estimated kinase activities in polyp tissue (E) and mucosal tissue (F). (G,H) KEA3 mean rank score difference of top 25 kinases estimated as enriched in polyps over mucosa. The scores of top 25 kinases based on mean rank score from KEA3 from polyps vs. mucosal tissue. Kinases with the largest differences in scores in polyp tissue over mucosal tissue are shown. ATR and CDK4 kinases are highlighted in red. (I,J) Expression analysis based on TCGA showing ATR and CDK4 values in normal vs. tumor tissue in rectal and colon adenocarcinoma. *, P<0.001. ATR, ataxia telangiectasia and Rad3-related kinase; CDK4; cyclin-dependent kinase 4; COAD, colon adenocarcinoma; READ, rectal adenocarcinoma; TCGA, The Cancer Genome Atlas; TPM, transcripts per million.

Next, from the statistically significantly upregulated proteins in polyps and mucosa we carried out a pathway analysis using Metascape (Figure 2C). Here we found significant enrichment in polyps of critical pathways accelerating cellular growth like DNA and RNA metabolic processes, proteins and pathways involved in translation, and regulation of RNA polymerase I, for example (Figure 2C). These datasets provide critical molecular insight into FAP.

Kinases are powerful potential therapeutic targets (43-45) and using phosphoproteomics is the ideal approach to infer upstream kinase activity in a cell (46-48). Along with proteome, we carried out a deep phosphoproteome analysis on the same set of samples and during the analysis, we discovered 4,233 phosphopeptides that were identified, and quantified. There were 352 phosphopeptides whose expression was statistically significantly increased in mucosa over polyps as is shown in Figure 2D. The five most upregulated phosphorylation sites in mucosa over polyps were Thr107 on neuromodulin, S322 on chromogranin-A, S708 on band 4.1-like protein 3, S225 on secretogranin-1, and S648 on band 4.1-like protein 3. Next, there were 182 phosphopeptides whose expression was statistically significantly increased in polyps over mucosa and as shown in Figure 2D. The five most upregulated phosphorylation sites in polyps over mucosa were S2013 on SON DNA and RNA Binding Protein, S624 on RNA-binding protein 5, T402 on Bcl-2-associated transcription factor 1, S621 on RNA-binding protein 5, and S39 on eukaryotic translation initiation factor 3 subunit C.

The next step in this analysis is to seek to determine which kinases might be active in polyps versus mucosa. To do this, a number of computational tools have been developed to infer the kinase activity in a cell using omics datasets, like phosphoproteomics. Here we carried out our preliminary analysis using kinase enrichment analysis 3 (KEA3) (38), which infers upstream kinases from substrates overrepresented in a list. KEA3 (38) provides a MeanRank Score, for example, for all known human kinases from a given dataset. By carrying this out for both polyps (Figure 2E) and mucosa (Figure 2F) we have a ranked list of all inferred kinases from each cell type. From this we then subtracted the ranks of each kinase in each dataset from the other and looked for the differences in ranks between each dataset (Figure 2G,2H). This approach then suggests that specific kinases were more active in polyps versus mucosa. Our preliminary data analysis suggests that kinases important in cancer like CDK4, CDK7 (49), CDC7 (50), and ATR (51,52), are more active in polyps than mucosa. The Cancer Genome Atlas (TCGA) data analysis of expression pattern of ATR and CDK4 further revealed higher levels of mRNA expression in both rectal and colon adenocarcinoma (Figure 2I,2J) indicating that increases seen in FAP parallel what is observed in CRC.

Regional differences in ATR and CDK4 kinases and effect of blocking these kinases on spheroid growth and colony formation

When we looked at ATR and CDK4 protein levels in the polyps, we observed relative abundance in both the ascending/transverse and descending colon polyps to be higher than normal mucosa (Figure 3A-3C). We substantiated these findings via immunohistochemical staining, wherein, both ATR and CDK4 levels were higher in the ascending/transverse vs. descending colon polyps (Figure 3D). Using Apc^Min/+ mice as a prototypic mouse model of polyposis, we further substantiated higher CDK4 and ATR staining in the adenomas compared to normal mucosa (Figure 3E). Since we have reported significant enrichment of Fn in the polyps (53), we generated CM by growing Fn anaerobically followed by centrifugation and 0.22 µm filtration (Figure 3F). HCT116 colon cancer cells were incubated with FnCM (1:1,000 dilution; 10.476 mg/mL protein) for 24 or 48 h. Following western blotting, we discovered higher relative abundance of both ATR and CDK4 levels at 24 and particularly at 48 h in response to FnCM treatment, compared to control (Figure 3G,3H). To test if blocking CDK4/6 or ATR through palbociclib (CDK4/6i) or ceralasertib (ATRi) affected either colony formation or spheroid assay, we seeded six-well plates with 500 viable cells per well, treated with CDK4/6i or ATRi either as monotherapy or in combination (25 µM) in 10% FBS containing DMEM for 48 h. Following removal of the inhibitor-containing media, the cells were incubated for an additional 10 days. The colonies obtained were fixed in formalin, stained with crystal violet and counted. The number of colonies following treatment was compared with the untreated cells. Both inhibitors either alone or in combination, significantly inhibited colony formation (Figure 3I,3J). Next, the HCT116 cells were cultured in DMEM supplemented with appropriate growth factors at low densities (5,000 cells/mL) in six-well low adhesion plates. The cells were treated with CDK4/6i or ATRi either as monotherapy or in combination (25 µM) and after 5 days, the number and size of colonospheres were determined. As is depicted in Figure 3K, both inhibitors significantly blocked spheroid formation. To further substantiate the inhibitory role of ATR, we performed both colony formation and spheroid assay using VE-821, a potent competitive inhibitor of ATR with a half-maximal inhibitory concentration (IC50) of 26 nM (54). Both colony formation as well as spheroid growth were significantly inhibited by VE-821 (Figure 3L-3O). These findings clearly demonstrate that targeting ATR and CDK4/6 could be a viable strategy to prevent or mitigate polyposis in FAP.

Figure 3 ATR and CDK4 expression in FAP. (A-C) Western blots (A) and bar graphs (B,C) showing relative abundance of ATR and CDK4 protein in NM or polyps (P1–P3). (D,E) Representative images showing ATR and CDK4 staining in the polyps vs. the normal mucosa (D) and in normal vs. the polyps of Apc^Min/+ mice. Scale bars =200 mm (D) and 100 mm (E). Yellow dashed boxes indicate representative regions of interest highlighting areas of increased ATR and CDK4 immunostaining. (F) Fn was grown anaerobically and CM was collected for infection of HCT116 cells. (G,H) Western blots (G) and bar graphs (H) showing the effect of FnCM treatment on ATR and CDK4 levels after 24 or 48 h. β-actin was used as loading control (n=3 technical replicates). (I,J) Colony formation assay. Six-well plates with 500 viable cells per well were treated with CDK4/6i or ATRi either as monotherapy or in combination (25 µM) in 10% FBS containing DMEM for 48 h. The colonies, after 10 days were fixed in formalin, stained with crystal violet (I) and counted (J) (P values as indicated; n=3 independent experiments). (K) Effect of CDK4/6i or ATRi either as monotherapy or in combination on spheroid growth. Representative images of colony morphology under control, ATRi, CDK4/6i, and combination treatment conditions (phase-contrast microscopy; no staining). (L-O) Effect of ATRi VE-821 on colony formation (L) and spheroid assay. (M) Colony formation assay plates fixed and stained with crystal violet. Spheroid assay showing representative images of colonospheres (phase-contrast microscopy; no staining). Bar graphs showing number of colonies (N) and spheroids (O) (*, P<0.05; **, P<0.001; n=3 technical replicates). ATR, ataxia telangiectasia and Rad3-related kinase; ATRi, ATR inhibitor; CDK4; cyclin-dependent kinase 4; CDK4/6i, CDK4/6 inhibitor; DMEM, Dulbecco’s modified Eagle medium; FAP, familial adenomatous polyposis; FBS, fetal bovine serum; Fn, Fusobacterium nucleatum; NM, normal mucosa.

Differential oxidative DNA damage in proximal vs. distal colon accompanies polyp growth in FAP

Prior publications have revealed that E. coli containing the polyketide synthase (pks) genotoxic island induces DNA damage in vitro and in vivo along with colon tumorigenesis in AOM-treated IL-10 KO mice (55), whereas enterotoxigenic Bacteroides fragilis (ETBF) induces colon tumorigenesis in Apc^Min/+ mice (56). γH2AX immunohistochemistry revealed significantly enhanced DNA damage in the colon epithelial cells of AOM mice co-colonized with pks⁺E. coli and ETBF compared to mono-colonized (pks⁺E. coli or ETBF) mice (56). Given our findings of Fusobacterium presence in the polyps of FAP patients (53), we hypothesized that presence of Fusobacterium can help facilitate a genotoxic environment wherein, single strand break (SSB), or double strand break (DSB) in the DNA can directly lead to neoplastic growth. As proof of concept, we incubated HCT116 cells with FnCM (1:1,000 dilution; 10.476 mg/mL protein) for 48 h. γH2AX immunocytochemistry revealed significantly enhanced DNA damage in the cells treated with FnCM compared to control (Figure 4A,4B). To see if similar trend is seen in vivo, we next stained the entire repertoire of normal mucosa or proximal and distal colon polyps (Figure 4C-4E) with antibodies against both 8-OxoG and gH2AX. As is depicted in Figure 4F, we discovered significant differences in staining of these two markers with ascending colon showing highest staining followed by transverse and descending. Moreover, while descending colon exhibited positive 8-OxoG staining representing SSB, gH2AX staining was either minimal or absent suggesting that DNA damage did not extend beyond SSB in the descending colon (Figure 4F). These results were consistent across diverse patient population and correlated with polyp size (Figure 5). Despite this being a consistent feature of the colon cancer with defects in MMR genes promoting MSI, this has never been reported in FAP and therefore represents a novel finding.

Figure 4 Effect of FnCM on DNA damage. HCT116 cells were exposed to PBS (control) or FnCM for 48 h and cells were stained with antibody to gH2AX. (A) High powered image showing gH2AX staining with mono (m) and polyclonal (p) antibodies. (B) Foci showing mean intensity in the two groups. Red line indicates median; P<0.001, Student’s t-test. (C-E) Representative images showing H&E staining of normal mucosa and ascending or descending polyps (scale bar =200 mm). (F) Immunofluorescence for 8-OxoG (green) and immunohistochemistry for gH2AX (brown) in different regions of the colon from a single FAP patient. DAPI, nuclei; scale bar =150 µm (n=3 technical replicates). CM, conditioned media; DAPI, 4’,6-diamidino-2-phenylindole; FAP, familial adenomatous polyposis; Fn, Fusobacterium nucleatum; gH2AX, phosphorylated histone H2A.X; H&E, hematoxylin and eosin; PBS, phosphate-buffered saline.

Figure 5 Regional differences in double strand breaks in FAP. Immunohistochemistry for gH2AX (brown) in different regions of the colon from multiple FAP patients. DAPI, nuclei; scale bar =150 µm. Lower panel: please note differences in gH2AX staining in a large polyp from each site. Scale bar =150 µm (n=3 technical replicates). DAPI, 4’,6-diamidino-2-phenylindole; FAP, familial adenomatous polyposis; gH2AX, phosphorylated histone H2A.X.

Immune profiling in proximal and distal colon polyps

Since several bacteria, including Fn (57), Escherichia coli (55), and ETBF (58), have been implicated in triggering CRC by altering the immune response and based on a recent report wherein loss of resident memory T cells and gdT cells in mucosal tissue of patients with FAP correlated with intestinal microbial dysbiosis (59), we performed the flow cytometry to isolate immune cells in either control mucosa or polyps. Figure S2 is a prototypic example of flow cytometry on frozen samples. For gating, however, we utilized fresh biopsies to establish viability before staining cells with antibodies to detect CD4⁺ or CD8⁺ cells and separated CD103⁺ cells as a subset of CD4 cells. As is depicted in Figure 6, CD103⁺ cells were present at higher levels in the ascending (Figure 6D) or descending (Figure 6F) colons polyps than normal mucosa (Figure 6C,6E) with comparable levels in the cecal polyps (Figure 6A,6B). Interestingly, polyps from descending colon (Figure 6F) exhibited the highest frequency of CD103⁺ cells compared to polyps from ascending colon (Figure 6D,6G). Given recent discovery of presence of CD4⁺ TRMs in various mucosal tissues including small intestine (60-63) wherein, these cells protect hosts against invading pathogens and based on discrete distribution of CD103⁺ cells in ascending vs. the descending colon (Figure 6C-6G), we stained the paraffin sections from the two colons with antibodies against LPS and LTA representing lipopolysaccharide and lipoteichoic acid to identify Gram negative and Gram positive bacteria, respectively. As is shown in Figure 6H and consistent with earlier report about predominance of bacterial biofilms in the proximal but not distal colons of FAP patients (56), we discovered significantly higher staining of both LPS and LTA in the ascending but not descending colon polyps indicating that elevated levels of CD103⁺ TRM in the distal colon may be protective.

Figure 6 Spectral flow cytometry of tissue biopsies revealed differences in immune cell frequencies. (A-F) Gating strategy for flow cytometry of normal mucosa or polyps. Viable singlet lymphocytes were evaluated for CD4 and CD8. CD4 cells were subsequently analyzed for CD103 in ascending and descending polyps. (G) Percent of CD103 subsets were plotted in each group. (H) Representative immunofluorescence for LPS and LTA, in the ascending and descending colons of an FAP patient. Scale bar =250 mm (n=3 technical replicates). FAP, familial adenomatous polyposis; LPS, lipopolysaccharide; LTA, lipoteichoic acid; NM, normal mucosa.

Growth of polyp biopsies as organoids

A complex three-dimensional (3D) structure that facilitates tissue-specific functions exists at mucosal surfaces lining the gastrointestinal tract. To mimic 3D microenvironment in vitro, we next isolated cells from polyps from different regions of the colon and grew them as organoids. Intriguingly, we discovered regional differences in organoid growth as is depicted in Figure 7. A varied growth pattern led to differences in the sizes of cecal/ascending/transverse and descending polyp organoids (Figure 7A,7B). Specifically, the area of transverse colon-derived polyp organoid was significantly higher than cecal or descending colon organoids (Figure 7C; P<0.0001). Cecal and transverse colon-derived polyp organoids also exhibited greater crypt budding phenomenon compared to descending colon (Figure 7D). To see what may be fueling this growth, we performed real-time PCR and looked at the expression profiles of stem cell markers such as CD44, CD133, Lgr5 and BMI-1, respectively. Epithelial cell adhesion molecule (EPCAM) was used to establish the purity of the isolated cell population. As is shown in Figure 7E, we discovered that cecal and transverse colon polyps showed higher stem cell marker expression consistent with differences in their growth as organoids and their budding characteristics. We can glean from these studies that regional differences in organoid growth exist in the same FAP subject and are fueled by expression levels of various stem cell markers.

Figure 7 Growth of polyps as organoids. (A,B) Phase contrast images demonstrating organoid growth at day 10 and day 30 post sample collection. Insets (blue boxes) represent higher magnification views (20×) of the highlighted regions. (C,D) Bar diagram showing area of organoid (C) and budding vs. total organoids ratio (D) of these organoids. Area of transverse colon derived polyp is significantly higher than cecal and descending colon derived polyp (****, P<0.0001). Cecal and transverse colon derived polyp showed significantly higher budding crypt compared to descending colon (**, P<0.004). (E) qPCR analysis demonstrated higher stem cell marker expression in cecal, and transverse colon compared to descending colon (NS, not significant; *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001). qPCR, quantitative polymerase chain reaction.

Discussion

CRC remains a formidable challenge especially given the emerging increased incidence in the younger patients. Our study sought to define the molecular landscape of adenomatous polyps in pediatric patients with FAP and to delineate the proteomic and phosphoproteomic differences between polyp and non-polyp mucosa. Our findings provide insights into the molecular mechanisms driving polyp progression and, by extension, the earliest stages of colorectal tumorigenesis in young patients.

Key findings

We have demonstrated distinct proteomic and phosphoproteomic profiles between polyp and non-polyp mucosa. We have identified hundreds of differentially expressed proteins between polyp and adjacent mucosa. The overarching theme suggests a pattern of over-expression of proteins involved in DNA metabolism and cellular growth in polyps compared to mucosa, and conversely, enrichment of proteins involved in cellular differentiation and homeostasis in mucosa compared to polyp tissue. These findings are aligned to earlier observations on differences in protein expression between polyp and non-polyp mucosa, highlighting the early molecular changes that favor proliferation in adenomatous polyps (64,65).

Strengths and limitations

In contrast with related studies, the colectomy derived specimens included in our study were obtained from colon not subjected to preoperative bowel cleansing regimen that has been shown to alter the microbiologic signature of the colon and influence observations (66). Despite these strengths, our study has notable limitations. The very small sample size (n=3) reflects the rarity of FAP overall and the refinement of pediatric patients undergoing colectomy. This does not detract from the consequence that our multi-omics observations, including regional differences in kinase activity, immune signatures, stemness pathways, and microbiome associations, have to be interpreted as exploratory and hypothesis-generating rather than definitive. In addition, despite our integrated proteomic, phosphoproteomic, immunologic, and organoid assays providing a cohesive framework for understanding regional polyp heterogeneity, these are in large part associative datasets necessitating further mechanistic validation, ahead of needs to be exerted when making causal inferences regarding microbiome epithelial interactions, DNA-damage signaling, or specific kinase-driven mechanisms. Relatedly, ATR/CDK4 and other pathway alterations identified in this work should be framed as potential, rather than confirmed therapeutic targets. Together, these limitations underscore the need for future studies with larger cohorts, longitudinal sampling, and targeted mechanistic experiments to validate and extend these preliminary observations.

Explanations of findings

Our study underscores the central role of kinase signaling in polyp growth and progression. While direct studies on ATR’s role in adenomatous polyps are limited, the published literature supports the involvement of CDK4 in polyp growth (67,68). We observed that both kinases are more active in polyps suggesting a role in driving polyp growth and tumorigenesis. Our observation of colorectal regional differences in the expression of kinase activity is particularly intriguing. We observed higher ATR and CDK4 expression in the right compared with left colon polyps.

These observations suggest regional differences in growth and oncogenic stressors, potentially driven by the tumor/tissue microenvironment, including oxidative stress, local microbiome, and immune dysregulation. The pathogenetic relevance of these kinases was underscored by inhibition experiments where targeted blockade of ATR or CDK4 significantly reduced polyp derived colony formation and spheroid growth. In turn, this suggests that kinase inhibitors targeting ATR or CDK4 could serve as chemopreventive or therapeutic agents in pediatric FAP, mitigating polyp burden before malignant transformation, and thus may play a role in delaying or obviating colectomy.

The interplay between microbial dysbiosis and oxidative DNA damage emerged as a significant contributor to polyp progression. We observed increased presence of Fn in polyps coupled with increased markers of DNA strand breaks (γH2AX) and oxidative stress (8-OxoG). Interestingly, proximal (right-sided) polyps showed a higher burden of double-strand DNA breaks in contrast to distal (left-sided) polyps which exhibited more single-strand breaks. This novel finding suggests that site-specific oxidative stress may drive regional variations in polyp progression and malignancy risk, potentially influenced by differences in bacterial biofilms or immune responses. In an earlier longitudinal study in pediatric patients with FAP, we reported that the relative abundance of Fusobacteria was higher in polyps compared to non-polyp and stool specimens over time (53). Fn is emerging as a potential driver in several different cancer types including CRC and across different populations with putative pro-tumorigenic mechanisms including activation of cellular proliferative pathways, tumorigenic alterations in the inflammatory microenvironment, immune evasion and chemoresistance through suppression of miRNA (69). Fn has been shown to induce DNA damage and tumor growth in Apc^Min/+ mice through FadA-dependent activation of the E-cadherin/β-catenin pathway (70) suggesting that in FAP, Fn effectively behaves as an accelerant superimposed on the underlying genetic defect.

TRMs are involved in cytotoxic pathways in the tumor microenvironment (71) and are an independent prognostic determinant in CRC (72). A right to left gradient in TRMs has been observed in CRC putatively related to the MSI characteristics of the tumor. In contrast in our study TRMs were enriched in left colon polyps whereas right sided colon polyps displayed a relative depletion of these protective immune subsets. This raises the possibility of regional differences in immune surveillance impacting differences in polyp progression and relatedly, the opportunity for chemopreventive strategies targeting local immune responses in this population.

Another key aspect of our study were the observed differences in growth of polyp-derived organoids. Organoids derived from the right colon polyps showed faster growth and crypt budding when compared to left colon polyps suggesting regional differences in stem cell activity. This was further supported by the observed increased expression of stem cell markers (CD44, CD133, Lgr5, and BMI-1) in right-sided polyps, indicating a more stem-like phenotype in these regions. This aligns with the observation that right-sided CRCs are often more aggressive, mucinous, and microsatellite unstable (MSI-high), features that may reflect a stem cell-like state (6). These findings underscore the profound differences in polyp biology across the colon, with proximal polyps potentially having a higher risk of transformation due to a greater reservoir of proliferative, undifferentiated cells.

Implications and actions needed

Thus, our findings have important clinical and translational implications. First, the identification of differentially expressed proteins and kinases provides potential biomarkers for early or non-invasive polyp detection, particularly in high-risk individuals like those with FAP. Second, our observations suggest ATR and CDK4 as promising therapeutic targets, with kinase inhibitors showing potential to suppress polyp growth. Third, our data reinforces the growing recognition of the microbiome as a key driver of adenomatous polyp growth and colorectal tumorigenesis, suggesting that microbiome-targeted interventions, such as pre- and probiotics, or antibiotics, could play a role in polyp prevention in this population. Thus, larger, region-stratified pediatric FAP cohorts are required to validate the observed gradients in DNA damage, kinase activation, microbial biofilms, and immune surveillance. Preclinical investigation into ATR/CDK4 inhibition and microbiome-targeted strategies in patient-derived organoids and animal models should be prioritized as a focus for chemoprevention trials.

Conclusions

Our study provides novel insights into the regional-biologic heterogeneity of polyps in FAP, revealing distinct proteomic, kinase, immune, and stem cell features across different colonic regions. These findings not only advance our understanding of FAP-associated polyp growth and progression but also highlight potential targets for polyp-burden surveillance and chemopreventive strategies. By integrating molecular, microbial, and immune profiling, we may move closer to precision prevention of CRC, ultimately improving outcomes for young people at high risk including with FAP.

Acknowledgments

The authors would like to acknowledge Dr. Julie Broski, Medical Writer, Department of Surgery, for her skills and dedication towards formatting of this manuscript.

Footnote

Reporting Checklist: The authors have completed the MDAR and ARRIVE reporting checklists. Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-138/rc

Data Sharing Statement: Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-138/dss

Peer Review File: Available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-138/prf

Funding: This work was supported by the IDeA National Resource for Quantitative Proteomics for the generation of proteomic and phosphoproteomic data, funded by the National Institute of General Medical Sciences (No. R24GM137786). The work presented here was also supported in part by pilot funds from the University of Kansas Cancer Center.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tgh.amegroups.com/article/view/10.21037/tgh-2025-138/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics board of Children’s Mercy Hospital (IRB Nos. 15020066 and 00001861) and informed consent was obtained from all individual participants or their legal guardians. Animal experiments were performed under a project license (No. 23-10-353) granted by University of Kansas Medical Center in compliance with the National Institutes of Health guidelines for the care and use of animals.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Saraiva MR, Rosa I, Claro I. Early-onset colorectal cancer: A review of current knowledge. World J Gastroenterol 2023;29:1289-303. [Crossref] [PubMed]
2. Cho YA, Lee J, Oh JH, et al. Genetic Risk Score, Combined Lifestyle Factors and Risk of Colorectal Cancer. Cancer Res Treat 2019;51:1033-40. [Crossref] [PubMed]
3. Song M, Chan AT, Sun J. Influence of the Gut Microbiome, Diet, and Environment on Risk of Colorectal Cancer. Gastroenterology 2020;158:322-40. [Crossref] [PubMed]
4. Alzahrani SM, Al Doghaither HA, Al-Ghafari AB. General insight into cancer: An overview of colorectal cancer Mol Clin Oncol 2021;15:271. (Review). [Crossref] [PubMed]
5. Nguyen LH, Goel A, Chung DC. Pathways of Colorectal Carcinogenesis. Gastroenterology 2020;158:291-302. [Crossref] [PubMed]
6. Baran B, Mert Ozupek N, Yerli Tetik N, et al. Difference Between Left-Sided and Right-Sided Colorectal Cancer: A Focused Review of Literature. Gastroenterology Res 2018;11:264-73. [Crossref] [PubMed]
7. Ciepiela I, Szczepaniak M, Ciepiela P, et al. Tumor location matters, next generation sequencing mutation profiling of left-sided, rectal, and right-sided colorectal tumors in 552 patients. Sci Rep 2024;14:4619. [Crossref] [PubMed]
8. Salem ME, Weinberg BA, Xiu J, et al. Comparative molecular analyses of left-sided colon, right-sided colon, and rectal cancers. Oncotarget 2017;8:86356-68. [Crossref] [PubMed]
9. Li Q, Geng S, Luo H, et al. Signaling pathways involved in colorectal cancer: pathogenesis and targeted therapy. Signal Transduct Target Ther 2024;9:266. [Crossref] [PubMed]
10. Ngoi NYL, Peng G, Yap TA. A Tale of Two Checkpoints: ATR Inhibition and PD-(L)1 Blockade. Annu Rev Med 2022;73:231-50. [Crossref] [PubMed]
11. Serrano-Heras G, Cuenca-López MD, Montero JC, et al. Phospho-kinase profile of colorectal tumors guides in the selection of multi-kinase inhibitors. Oncotarget 2015;6:31272-83. [Crossref] [PubMed]
12. Li P, Huang D. Targeting the JAK-STAT pathway in colorectal cancer: mechanisms, clinical implications, and therapeutic potential. Front Cell Dev Biol 2024;12:1507621. [Crossref] [PubMed]
13. Ceccaldi R, Rondinelli B, D'Andrea AD. Repair Pathway Choices and Consequences at the Double-Strand Break. Trends Cell Biol 2016;26:52-64. [Crossref] [PubMed]
14. Martin OCB, Frisan T. Bacterial Genotoxin-Induced DNA Damage and Modulation of the Host Immune Microenvironment. Toxins (Basel) 2020;12:63. [Crossref] [PubMed]
15. Rodgers K, McVey M. Error-Prone Repair of DNA Double-Strand Breaks. J Cell Physiol 2016;231:15-24. [Crossref] [PubMed]
16. Zhou X, Kandalai S, Hossain F, et al. Tumor microbiome metabolism: A game changer in cancer development and therapy. Front Oncol 2022;12:933407. [Crossref] [PubMed]
17. Galasso L, Termite F, Mignini I, et al. Unraveling the Role of Fusobacterium nucleatum in Colorectal Cancer: Molecular Mechanisms and Pathogenic Insights. Cancers (Basel) 2025;17:368. [Crossref] [PubMed]
18. Rubinstein MR, Wang X, Liu W, et al. Fusobacterium nucleatum promotes colorectal carcinogenesis by modulating E-cadherin/β-catenin signaling via its FadA adhesin. Cell Host Microbe 2013;14:195-206. [Crossref] [PubMed]
19. Wang S, Liu Y, Li J, et al. Fusobacterium nucleatum Acts as a Pro-carcinogenic Bacterium in Colorectal Cancer: From Association to Causality. Front Cell Dev Biol 2021;9:710165. [Crossref] [PubMed]
20. Pignatelli P, Nuccio F, Piattelli A, et al. The Role of Fusobacterium nucleatum in Oral and Colorectal Carcinogenesis. Microorganisms 2023;11:2358. [Crossref] [PubMed]
21. Queen J, Cing Z, Minsky H, et al. Fusobacterium nucleatum is enriched in invasive biofilms in colorectal cancer. NPJ Biofilms Microbiomes 2025;11:81. [Crossref] [PubMed]
22. Mezheyeuski A, Micke P, Martín-Bernabé A, et al. The Immune Landscape of Colorectal Cancer. Cancers (Basel) 2021;13:5545. [Crossref] [PubMed]
23. Zheng Z, Wieder T, Mauerer B, et al. T Cells in Colorectal Cancer: Unravelling the Function of Different T Cell Subsets in the Tumor Microenvironment. Int J Mol Sci 2023;24:11673. [Crossref] [PubMed]
24. Bai Z, Zhou Y, Ye Z, et al. Tumor-Infiltrating Lymphocytes in Colorectal Cancer: The Fundamental Indication and Application on Immunotherapy. Front Immunol 2021;12:808964. [Crossref] [PubMed]
25. Santos AJM, Lo YH, Mah AT, et al. The Intestinal Stem Cell Niche: Homeostasis and Adaptations. Trends Cell Biol 2018;28:1062-78. [Crossref] [PubMed]
26. Kim KM, Calabrese P, Tavaré S, et al. Enhanced stem cell survival in familial adenomatous polyposis. Am J Pathol 2004;164:1369-77. [Crossref] [PubMed]
27. Hyer W, Cohen S, Attard T, et al. Management of Familial Adenomatous Polyposis in Children and Adolescents: Position Paper From the ESPGHAN Polyposis Working Group. J Pediatr Gastroenterol Nutr 2019;68:428-41. [Crossref] [PubMed]
28. Keswani RN. Cold Snare Polypectomy: Techniques and Applications. Clin Gastroenterol Hepatol 2020;18:42-4. [Crossref] [PubMed]
29. Yusuf K, Roy BC, Hauser WL, et al. DCLK1 isoform (DCLK1-S) as a critical player in promoting inflammation, tissue remodeling, and EMT in mouse models of colitis. PLoS Pathog 2025;21:e1013360. [Crossref] [PubMed]
30. Sato T, Stange DE, Ferrante M, et al. Long-term expansion of epithelial organoids from human colon, adenoma, adenocarcinoma, and Barrett's epithelium. Gastroenterology 2011;141:1762-72. [Crossref] [PubMed]
31. Prianichnikov N, Koch H, Koch S, et al. MaxQuant Software for Ion Mobility Enhanced Shotgun Proteomics. Mol Cell Proteomics 2020;19:1058-69. [Crossref] [PubMed]
32. Nesvizhskii AI, Keller A, Kolker E, et al. A statistical model for identifying proteins by tandem mass spectrometry. Anal Chem 2003;75:4646-58. [Crossref] [PubMed]
33. Graw S, Tang J, Zafar MK, et al. proteiNorm - A User-Friendly Tool for Normalization and Analysis of TMT and Label-Free Protein Quantification. ACS Omega 2020;5:25625-33. [Crossref] [PubMed]
34. Webb-Robertson BJ, Matzke MM, Jacobs JM, et al. A statistical selection strategy for normalization procedures in LC-MS proteomics experiments through dataset-dependent ranking of normalization scaling factors. Proteomics 2011;11:4736-41. [Crossref] [PubMed]
35. Thurman TJ, Washam CL, Alkam D, et al. proteoDA: a package for quantitative proteomics. Journal of Open Source Software 2023;8:5184.
36. Ritchie ME, Phipson B, Wu D, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015;43:e47. [Crossref] [PubMed]
37. Zhou Y, Zhou B, Pache L, et al. Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 2019;10:1523. [Crossref] [PubMed]
38. Kuleshov MV, Xie Z, London ABK, et al. KEA3: improved kinase enrichment analysis via data integration. Nucleic Acids Res 2021;49:W304-16. [Crossref] [PubMed]
39. Perez-Riverol Y, Bandla C, Kundu DJ, et al. The PRIDE database at 20 years: 2025 update. Nucleic Acids Res 2025;53:D543-53. [Crossref] [PubMed]
40. Deutsch EW, Bandeira N, Perez-Riverol Y, et al. The ProteomeXchange consortium at 10 years: 2023 update. Nucleic Acids Res 2023;51:D1539-48. [Crossref] [PubMed]
41. Perez-Riverol Y, Xu QW, Wang R, et al. PRIDE Inspector Toolsuite: moving towards a universal visualization tool for proteomics data standard formats and quality assessment of ProteomeXchange datasets. Molecular & Cellular Proteomics 2016;15:305-17. [Crossref] [PubMed]
42. Mertins P, Tang LC, Krug K, et al. Reproducible workflow for multiplexed deep-scale proteome and phosphoproteome analysis of tumor tissues by liquid chromatography-mass spectrometry. Nat Protoc 2018;13:1632-61. [Crossref] [PubMed]
43. Butti R, Das S, Gunasekaran VP, et al. Receptor tyrosine kinases (RTKs) in breast cancer: signaling, therapeutic implications and challenges. Mol Cancer 2018;17:34. [Crossref] [PubMed]
44. Nagini S. Breast Cancer: Current Molecular Therapeutic Targets and New Players. Anticancer Agents Med Chem 2017;17:152-63. [Crossref] [PubMed]
45. Pascual J, Turner NC. Targeting the PI3-kinase pathway in triple-negative breast cancer. Ann Oncol 2019;30:1051-60. [Crossref] [PubMed]
46. Liu Z, Liu Y, Qian L, et al. A proteomic and phosphoproteomic landscape of KRAS mutant cancers identifies combination therapies. Mol Cell 2021;81:4076-4090.e8. [Crossref] [PubMed]
47. Maxwell MJ, Arnold A, Sweeney H, et al. Unbiased Proteomic and Phosphoproteomic Analysis Identifies Response Signatures and Novel Susceptibilities After Combined MEK and mTOR Inhibition in BRAF(V600E) Mutant Glioma. Mol Cell Proteomics 2021;20:100123. [Crossref] [PubMed]
48. Recasens A, Humphrey SJ, Ellis M, et al. Global phosphoproteomics reveals DYRK1A regulates CDK1 activity in glioblastoma cells. Cell Death Discov 2021;7:81. [Crossref] [PubMed]
49. Sava GP, Fan H, Coombes RC, et al. CDK7 inhibitors as anticancer drugs. Cancer Metastasis Rev 2020;39:805-23. [Crossref] [PubMed]
50. Melling N, Muth J, Simon R, et al. Cdc7 overexpression is an independent prognostic marker and a potential therapeutic target in colorectal cancer. Diagn Pathol 2015;10:125. [Crossref] [PubMed]
51. Bradbury A, Hall S, Curtin N, et al. Targeting ATR as Cancer Therapy: A new era for synthetic lethality and synergistic combinations? Pharmacol Ther 2020;207:107450. [Crossref] [PubMed]
52. Lecona E, Fernandez-Capetillo O. Targeting ATR in cancer. Nat Rev Cancer 2018;18:586-95. [Crossref] [PubMed]
53. Attard TM, Septer S, Lawson CE, et al. Microbiome insights into pediatric familial adenomatous polyposis. Orphanet J Rare Dis 2022;17:416. [Crossref] [PubMed]
54. Reaper PM, Griffiths MR, Long JM, et al. Selective killing of ATM- or p53-deficient cancer cells through inhibition of ATR. Nat Chem Biol 2011;7:428-30. [Crossref] [PubMed]
55. Arthur JC, Perez-Chanona E, Mühlbauer M, et al. Intestinal inflammation targets cancer-inducing activity of the microbiota. Science 2012;338:120-3. [Crossref] [PubMed]
56. Dejea CM, Fathi P, Craig JM, et al. Patients with familial adenomatous polyposis harbor colonic biofilms containing tumorigenic bacteria. Science 2018;359:592-7. [Crossref] [PubMed]
57. Kostic AD, Chun E, Robertson L, et al. Fusobacterium nucleatum potentiates intestinal tumorigenesis and modulates the tumor-immune microenvironment. Cell Host Microbe 2013;14:207-15. [Crossref] [PubMed]
58. Chung L, Thiele Orberg E, Geis AL, et al. Bacteroides fragilis Toxin Coordinates a Pro-carcinogenic Inflammatory Cascade via Targeting of Colonic Epithelial Cells. Cell Host Microbe 2018;23:203-214.e5. [Crossref] [PubMed]
59. Noble A, Durant L, Dilke SM, et al. Altered Mucosal Immune-Microbiota Interactions in Familial Adenomatous Polyposis. Clin Transl Gastroenterol 2022;13:e00428. [Crossref] [PubMed]
60. Beura LK, Fares-Frederickson NJ, Steinert EM, et al. CD4(+) resident memory T cells dominate immunosurveillance and orchestrate local recall responses. J Exp Med 2019;216:1214-29. [Crossref] [PubMed]
61. Steinfelder S, Rausch S, Michael D, et al. Intestinal helminth infection induces highly functional resident memory CD4(+) T cells in mice. Eur J Immunol 2017;47:353-63. [Crossref] [PubMed]
62. Romagnoli PA, Fu HH, Qiu Z, et al. Differentiation of distinct long-lived memory CD4 T cells in intestinal tissues after oral Listeria monocytogenes infection. Mucosal Immunol 2017;10:520-30. [Crossref] [PubMed]
63. Kumar BV, Ma W, Miron M, et al. Human Tissue-Resident Memory T Cells Are Defined by Core Transcriptional and Functional Signatures in Lymphoid and Mucosal Sites. Cell Rep 2017;20:2921-34. [Crossref] [PubMed]
64. Wiśniewski JR, Duś-Szachniewicz K, Ostasiewicz P, et al. Absolute Proteome Analysis of Colorectal Mucosa, Adenoma, and Cancer Reveals Drastic Changes in Fatty Acid Metabolism and Plasma Membrane Transporters. J Proteome Res 2015;14:4005-18. [Crossref] [PubMed]
65. Shao Y, Xu K, Zheng X, et al. Proteomics profiling of colorectal cancer progression identifies PLOD2 as a potential therapeutic target. Cancer Commun (Lond) 2022;42:164-9. [Crossref] [PubMed]
66. Nagata N, Tohya M, Fukuda S, et al. Effects of bowel preparation on the human gut microbiome and metabolome. Sci Rep 2019;9:4042. [Crossref] [PubMed]
67. Su T, Washington MK, Ness RM, et al. Comparison of biomarker expression between proximal and distal colorectal adenomas: The Tennessee-Indiana Adenoma Recurrence Study. Mol Carcinog 2017;56:761-73. [Crossref] [PubMed]
68. Grady WM, Willis JE, Trobridge P, et al. Proliferation and Cdk4 expression in microsatellite unstable colon cancers with TGFBR2 mutations. Int J Cancer 2006;118:600-8. [Crossref] [PubMed]
69. Brennan CA, Garrett WS. Fusobacterium nucleatum - symbiont, opportunist and oncobacterium. Nat Rev Microbiol 2019;17:156-66. [Crossref] [PubMed]
70. Guo P, Tian Z, Kong X, et al. FadA promotes DNA damage and progression of Fusobacterium nucleatum-induced colorectal cancer through up-regulation of chk2. J Exp Clin Cancer Res 2020;39:202. [Crossref] [PubMed]
71. McAllister SS, Weinberg RA. The tumour-induced systemic environment as a critical regulator of cancer progression and metastasis. Nat Cell Biol 2014;16:717-27. [Crossref] [PubMed]
72. Talhouni S, Fadhil W, Mongan NP, et al. Activated tissue resident memory T-cells (CD8+CD103+CD39+) uniquely predict survival in left sided "immune-hot" colorectal cancers. Front Immunol 2023;14:1057292. [Crossref] [PubMed]