Indole Profiles Define Disease-Specific Virulence in Fusobacterium nucleatum Clinical Isolates

Clinical isolates of Fusobacterium nucleatum display strain-specific virulence and modulation by indole derivatives | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Clinical isolates of Fusobacterium nucleatum display strain-specific virulence and modulation by indole derivatives View ORCID Profile Colin Scano , View ORCID Profile Ankan Choudhury , View ORCID Profile Macarena G. Rojo , Jessalyn Hawkins , Lauren Crowhurst , Greg Zaharas , View ORCID Profile Ramon Lavado , View ORCID Profile Leigh Greathouse doi: https://doi.org/10.1101/2025.05.16.654558 Colin Scano 1 Biology Department, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Colin Scano Ankan Choudhury 2 Robbins College of Health and Human Sciences, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ankan Choudhury Macarena G. Rojo 3 Environmental Science, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Macarena G. Rojo Jessalyn Hawkins 1 Biology Department, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site Lauren Crowhurst 1 Biology Department, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site Greg Zaharas 1 Biology Department, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site Ramon Lavado 3 Environmental Science, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Ramon Lavado Leigh Greathouse 1 Biology Department, Baylor University 2 Robbins College of Health and Human Sciences, Baylor University Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Leigh Greathouse For correspondence: leigh_greathouse{at}baylor.edu Abstract Full Text Info/History Metrics Supplementary material Preview PDF ABSTRACT Pathogenic bacteria adapt to distinct disease environments, but whether these adaptations create therapeutic vulnerabilities remains unclear. Fusobacterium nucleatum has emerged as a key microbial player in colorectal cancer (CRC), yet its strain-specific virulence mechanisms remain poorly defined. In this pilot study of 16 clinical F. nucleatum isolates from CRC patients (n=6), Crohn’s disease patients (n=6), colon of healthy individuals (n=3), and an oral lesion (n=1), a subset of CRC-derived strains produced 3–4-fold higher levels of endogenous indole. Exogenous indole treatment differentially affected growth and biofilm formation, with some strains increasing biofilm despite growth inhibition. Notably, sensitivity to exogenous indole was independent of endogenous production and revealed that isolate 7-1 (EAVG002), a member of the tumorigenic Fna C2 clade, was uniquely hypersensitive to I3CA- and IPA-mediated stress. Invasion assays further showed that indole and its derivatives (I3A, I3CA) reduced invasion of a highly indole-tolerant CRC-derived isolate (SB-CTX3Tcol3) into CRC cells by ∼50%, comparable to antibiotic treatment. Furthermore, in two CRC cell lines, exposure to indole or indole derivatives resulted in substantial variability in adherens junction and tight junction transcript levels, with I3A having the strongest effect on tight junction ( CLDN1, CLDN7 ) transcripts. Collectively, these findings reveal profound strain-level heterogeneity and indole derivative effects, highlighting vulnerabilities that could enable precision therapeutic targeting of pathogenic F. nucleatum populations within the host environment while preserving beneficial commensals. INTRODUCTION The discovery that Fusobacterium nucleatum (F. nucleatum) selectively colonizes colorectal cancer (CRC) ( 1 , 2 ) transformed our understanding of host-microbe interactions in cancer, revealing that pathogenic bacteria not only respond to disease environments but may actively drive oncogenesis. While these F. nucleatum populations are thought to originate from the oral cavity, it remains unclear whether disseminated strains retain their original virulence programs or undergo niche-specific adaptation to the tumor environment ( 3 ). Recent pangenomic analysis suggests that virulence determinants and functional clusters in F. nucleatum are largely strain-specific rather than subspecies-specific, highlighting the genomic plasticity of individual isolates to adapt to distinct niches and optimize fitness ( 4 ). F. nucleatum is selectively enriched in invasive bacterial biofilms associated with CRC ( 5 ), a pathogenic mechanism that may also be mirrored in inflammatory bowel and Crohn’s disease ( 6 , 7 ). These invasive bacterial biofilms are carcinogenic in mouse models and correlate with reduced E-cadherin expression and increased crypt cell proliferation, directly linking bacterial biofilm to disease progression ( 5 , 8 , 9 ). Though F. nucleatum ’s role in Crohn’s disease is still under investigation, CD-derived isolates—specifically the tumorigenic Fna C2 strain 7-1 (EAVG002)—exhibit invasive and pro-inflammatory phenotypes that mirror those of CRC-derived strains ( 7 , 10 , 11 ). Beyond basic biofilm formation, F. nucleatum deploys an arsenal of outer membrane proteins and adhesins (including RadD, Aid1, and FadA), alongside the major structural porin FomA, that mediate both polymicrobial interactions and host cell binding ( 12 – 17 ). Furthermore, adherence to host cells enables F. nucleatum to trigger pro-inflammatory signaling pathways. The best-characterized adhesin, FadA, binds to E-cadherin to facilitate invasion ( 18 ), activate β-catenin signaling ( 15 , 19 ), and exacerbate DSS-induced colitis ( 20 ). Similarly, FomA serves as a critical anchor for epithelial adherence and immune modulation ( 16 , 21 ). Yet a critical knowledge gap remains: what environmental signals regulate these diverse virulence programs in the diseased environment? Identifying mechanisms regulating niche adaptation, including biofilm formation, virulence factor expression, and host cell invasion, is central to understanding F. nucleatum’s pathogenicity ( 5 , 15 , 22 ). One candidate regulator of niche adaptation is indole, a chemical effector in enteric communities that modulates biofilm formation and virulence across various pathogens ( 23 ). While early studies indicated that indole enhances biofilm formation in the oral-derived F. nucleatum reference strain ATCC 25586, it remains unknown whether this response is conserved across clinical isolates ( 24 ). Recent meta-analyses of fecal microbiome data indicate that tryptophan biosynthesis and metabolism are significantly enriched in CRC tumor samples compared to healthy colon controls (Supplementary Figure 1) ( 25 – 27 ) and increased indole levels can contribute to increased gut inflammation ( 28 ). Beyond these broad shifts, recent evidence has confirmed that specific indole derivatives produced by F. nucleatum can drive pro-carcinogenic pathways ( 29 , 30 ). However, it remains unclear how the production of indole and its derivatives regulate F. nucleatum virulence, and whether these metabolites act as niche-specific signals. In this pilot investigation, we hypothesized that F. nucleatum isolates from distinct disease environments harbor unique adaptations manifesting as differential responses to indole. We specifically examined whether a strain’s phenotypic response to exogenous indole is independent of its endogenous production profiles, and how specific indole derivatives affected pathogenic behavior. Overall, our study identified strain-specific responses to indole and indole derivatives and identified metabolic vulnerabilities that may be exploited for selective therapeutic suppression. MATERIALS AND METHODS Bacterial Anaerobic Culture Conditions F. nucleatum isolates were cultured from glycerol stocks on anaerobically reduced Columbia blood agar (5% sheep blood) or in Columbia broth supplemented with 0.5 g/L L-cysteine, 5 mg/L hemin, and 10 µg/L vitamin K (CB). Cultures were incubated at 37°C for 48–72 hours in a Coy Laboratories anaerobic chamber (90% N 2 , 5% CO 2 , 5% H 2 ). Quantification of total indole production with Kovacs reagent Total indole production was quantified using Kovac’s reagent. Stationary phase cultures were adjusted to 0.3 OD 600 and inoculated into four media types: CB, Brain Heart Infusion (BHI), or CB supplemented with 5 mM tryptophan or 1% glucose. Supernatants collected at 12-hour intervals (up to 72 h) were reacted with an equal volume of Kovac’s reagent for 5 minutes. Absorbance (OD 560 ) was measured using a Sunrise Tecan reader, and concentrations were determined via a 0–2000 µM standard curve. Data represent three biological replicates per isolate across each media condition. Quantification of indole derivatives by HPLC Starter cultures were adjusted to 0.3 OD 600 and inoculated into CB for 48 h. Supernatants were collected via centrifugation and analyzed using an UltiMate 3000 HPLC system with an RS fluorescence detector (Thermo Fisher Scientific) according to previously described methods ( 31 , 32 ). Separation was performed on an Acclaim™ VANQUISH C-18 column (2.2 μm, 150 × 2.1 mm) at 30°C with a flow rate of 0.5 mL/min. The mobile phase utilized a linear gradient of eluent A (50 mM KH 2 PO 4 , pH 3.3) and eluent B (70:30 eluent A:acetonitrile): 20% B (0–2 min), 20% to 75% B (2–10 min), 100% B (10–15 min), and 20% B (15–20 min). Native fluorescence was monitored at excitation/emission wavelengths of 270/350 nm. Targeted compounds (indole (CAS# 120-72-9), IPA (CAS# 830-96-6), IAA (CAS# 87-51-4), I3A (CAS# 487-89-8), and I3CA (CAS# 771-50-6)) were identified via authentic standards, confirmed by LC-MS, and quantified by peak area integration. Method linearity and system suitability (retention time stability, resolution) were monitored throughout. Quantification of Biofilm Biomass Biofilm biomass was quantified in 96-well plates via crystal violet (CV) staining ( 24 ). Briefly, biofilms were washed, air-dried, and fixed with methanol (37°C, overnight). Fixed biofilms were stained with 0.1% CV for 1 minute, washed, and destained with 30% acetic acid. Absorbance (A 560 ) was measured using a Sunrise Tecan microplate reader following a 10-second orbital shake. Quantification of F. nucleatum biofilm over 72 hours For time-course analysis, isolate cultures were grown to stationary phase, adjusted to 0.3 OD 600 in CB, and used to inoculate six 96-well plates simultaneously. Two uninoculated wells per plate served as contamination and background controls. Plates were harvested at 12-hour intervals over 72 h; bacterial density was measured (OD 600 ) using a Cerillo Stratus reader prior to biofilm quantification as described previously. Six biological replicates were performed per isolate. Biofilm was normalized to the maximum value per timepoint. Quantification of Indole-Mediated Growth and Biofilm Alterations Growth kinetics of 16 isolates were monitored under anaerobic conditions for 72 h (OD 600 every hour) using a Cerillo Stratus reader. Isolates were treated with five indole derivatives (0.25–2 mM) in triplicate; controls included 0.1% v/v DMSO (vehicle), tryptophan, and chloramphenicol. Following the 72-h kinetic run, the same experimental plates were immediately processed for biofilm quantification as previously described. Statistical significance was determined by one-way ANOVA followed by Tukey’s HSD post-hoc test. Statistical Modeling of Indole Sensitivity and Correlation Analysis Growth (AUC) and biofilm mass were normalized relative to untreated vehicle controls (baseline = 1.0) for each isolate–indole combination (Supplementary Figure 5, 6, and 7) . For each isolate–indole combination, signed deviations from the baseline were calculated at each tested concentration (0.25, 0.5, 1.0, and 2.0 mM). To determine overall sensitivity, only the Maximum Signed Deviation (Δ) was retained for each pair; this was defined as the single largest absolute deviation across all concentrations, preserving its original positive or negative sign. Isolates were hierarchically clustered based on Max Δ score profiles across treatments using Euclidean distance and Ward.D2 linkage (Supplementary Figure 8) . Statistical significance of the Maximum Signed Deviation (Δ) relative to untreated control was assessed using one-way ANOVA with Benjamini–Hochberg false discovery rate (FDR) correction (p < 0.05, p < 0.01, p < 0.001). To quantify phenotypic response magnitude, growth and biofilm Δ scores were projected against each other and calculated as the Euclidean distance from the untreated origin (0,0): Responce Magnitude = √Δ Growth 2 + Δ Biofilm 2 . Hyper-responsive isolates were defined as those falling outside a 75% confidence ellipse. The relationship between endogenous indole production and exogenous indole sensitivity was evaluated using Spearman’s rank correlation (ρ) and linear regression. Bacterial RNA Isolation and virulence gene qPCR F. nucleatum isolates (initial OD 600 = 0.3) were treated in triplicate with 1 mM indole derivatives or 0.1% DMSO. After 72 h anaerobic incubation, cells were harvested (8,000×g, 5 min) and lysed using lysozyme (0.5 mg/mL), 0.1 M sodium acetate, and 1% SDS. RNA was isolated via acid-phenol:chloroform extraction at 65°C, purified using Phasemaker tubes, and treated with TURBO DNase. RNA integrity was confirmed by Nanodrop and Qubit assays. cDNA was synthesized from 500 ng of total RNA; no-RT controls confirmed the absence of gDNA. Expression of virulence and metabolic genes (tnaA, aid1, fomA, fadA, radD, (Supplementary Table 1) ) was quantified using SYBR Green chemistry on a QuantStudio 6 Flex system. Target gene expression was normalized to 16S rRNA and calculated using the 2 −ΔΔC T method relative to the vehicle control. Primer specificity was confirmed on genomic DNA. Phylogenetic Tree Construction To evaluate the relationship between phylogeny and indole-mediated transcriptional responses, a Maximum Likelihood (ML) tree was constructed using 16S rRNA sequences from 16 isolates (Supplementary Table 2) . Sequences were obtained from suppliers, previously published genomic data ( 33 ) or generated via Sanger sequencing of an ∼800 bp PCR amplicon (primers 8F/907R) following previously established protocols ( 34 ). Sequences were aligned using DECIPHER in R. An initial Neighbor-Joining tree was generated using phangorn and optimized via ML for topology, base frequencies, substitution rates, and gamma-distributed rate variation. The tree was rooted with F. nucleatum ATCC 25586 and converted to an ultrametric format for direct heatmap alignment using ggtree and aplot ( Figure 6A ) . Predictive Modeling and Statistical Validation Naive Bayes classifiers were developed to predict clinical origin (CRC, Crohn’s, or Healthy colon) based on indole-induced transcriptional signatures. We compared a quantitative “Fold Change” model (2 −ΔΔC T ) against a categorical “Expression Direction” model (thresholds: (Up: >1.25; Down: <0.75; No Change: 1.25–0.75)). Models were trained on a 70% and validated on a 30% test set. Data split was validated using 10-fold cross-validation (Supplementary Figure 10A). Performance was evaluated via Accuracy, AUC-ROC, and macro-averaged Precision-Recall curves ( Figure 6E ) . Model calibration was verified through posterior class probabilities (Supplementary Figure 9C) , multiclass Brier scores, and faceted calibration curves (Supplementary Figure 10B-C) . To ensure feature independence, we calculated Mutual Information (threshold <0.5; Supplementary Figure 9A) and Cramér’s V statistics (Supplementary Figure 9D) . Feature importance was determined by calculating the Area Under the Curve (AUC) for each gene-class pair to quantify discriminatory power ( Figure 6D ). Assessment of Mammalian Epithelial Integrity Caco-2 and HT-29 cells (gift from Dr. Christie Sayes, Baylor University) were maintained in DMEM supplemented with 10% FBS and 200 μg/mL penicillin-streptomycin at 37°C (5% CO 2 ). Cells were seeded in 12-well plates (0.1 x 10 6 /well) and grown to 95% confluence before treatment with 1 mM indole derivatives, 0.1% DMSO (vehicle), or 10 µg/mL LPS (positive control for barrier disruption) for 24 h. Total RNA was extracted and reverse-transcribed, with no-RT controls confirming absence of DNA. qPCR was performed using PowerUp SYBR Green on a QuantStudio 6 system with 250 nM primers (Supplementary Table 1) . Relative expression of junctional genes was quantified via the 2 -ΔΔCt method, normalized to β-actin and relative to the vehicle control (DMSO). Bacterial Invasion Assay Caco-2 cells (∼85% confluence) in 24-well plates were incubated in antibiotic- and serum-free DMEM containing 1 mM indole derivatives. Cells were infected with log-phase F. nucleatum (isolate SB-CTX3Tcol3) at an MOI of 50:1 for 2 or 4 h. Following infection, monolayers were washed three times with PBS and treated with 200 µg/mL penicillin-streptomycin for 1 h to eliminate extracellular bacteria. After additional washing, cells were lysed with sterile deionized water. Lysates were serially diluted and plated in triplicate on complete medium agar. CFU/mL was calculated after 48 h of anaerobic incubation at 37°C. Statistical Methods All statistical analyses were conducted in R (v4.5.1) using the tidyverse package for data curation and ggplot2 for visualization. Differences across multiple experimental groups were assessed using one-way Analysis of Variance (ANOVA). To control for family-wise error rates during multiple comparisons, ANOVA was followed by Tukey’s Honest Significant Difference (HSD) post-hoc test. Statistical significance was defined a priori at α = 0.05. Unless otherwise noted, quantitative data are presented as mean ± standard error of the mean (SEM). RESULTS Isolates exhibit strain-specific indole production profiles Using Kovac’s reagent to capture a broad profile of indole-containing metabolites ( 35 ), we quantified indole production across 16 clinical F. nucleatum isolates over a 72-hour period ( Figure 1A ) . Indole production generally increased after 36 hours, corresponding with the transition from exponential to stationary growth phase. While production was observed for all isolates, there was significant heterogeneity within the isolate groups. Notably, a subset of CRC-derived isolates (SB-CTX3Tcol3, SB-CTX47T, CC7_3JVN3C1, and CC2_3FMU1) exhibited a 3- to 4-fold increase in indole production relative to the other isolates ( Figure 1B ) . This heightened production was consistent across different growth media ( Supplementary Figure 2 ). These distinctions were less pronounced when grouped by subspecies, highlighting strain-to-strain differences over classical subspecies classification ( Figure 1C ) . These results demonstrate that, within this cohort of isolates tested, indole production may be strain-specific and partially influenced by clinical origin, independent of subspecies classification. Download figure Open in new tab Figure 1: Isolates exhibit strain-specific indole production profiles. Kinetic profile of indole concentration quantified via Kovac’s reagent assay. Values are mean ± SEM per isolate. (B, C) Differences in indole levels when isolates are grouped by disease association (B) and subspecies (C). Bars indicate median with IQR. Statistical analysis performed using adjusted ANOVA and Tukey’s HSD comparisons (****p < 0.0001; ***p < 0.001; **p < 0.01; *p < 0.05). Group sizes: CRC (n=6), Crohn’s (n=6), healthy colon (n=3), oral reference strain (ATCC 25586) (n=1). Subspecies groups: animalis (n=5), vincentii (n=6), nucleatum (n=4), polymorphum (n=1). Isolates exhibit strain-specific indole derivative profiles and metabolic preferences To characterize the specific metabolic outputs of indole derivatives by individual isolates of F. nucleatum , we employed targeted HPLC analysis of five key derivatives that were differentially abundant in CRC patient stool ( 36 ). While all targeted derivatives—indole, indole-3-aldehyde (I3A), indole-3-carboxylic acid (I3CA), indole-3-acetic acid (IAA), and indole-3-propionic acid (IPA)—were detected across the cohort, IPA levels remained below the limit of detection in some samples ( Figure 2A ; Supplementary Figure 3A) . Notably, a subset of CRC- and Crohn’s-derived isolates exhibited significantly higher production of indole, I3CA, and IAA compared to healthy colon-derived isolates (p < 0.05). Consistent with our earlier findings, subspecies classification did not account for these differences (Supplementary Figure 3B) . Correlation analysis and derivative-to-parent indole ratios revealed that isolates possess distinct metabolic signatures or preferences ( Figure 2B ) . Specifically, three CRC-derived isolates (SB-CTX3Tcol3, SB-CTX47T, and CC2_6JVN3) demonstrated a marked preference for I3A production, with I3A/indole ratios (0.52–0.89) substantially exceeding those of other clinical or reference strains ( Figure 2C ) . These data suggest that individual isolates derived from disease environments may harbor specific adaptations in indole derivative production. Download figure Open in new tab Figure 2: Isolates exhibit strain-specific indole derivative profiles and metabolic preferences. (A) HPLC quantification of indole and five derivatives in Columbia Broth spent media. Box plots illustrate the production of indole, I3A, I3CA, IAA, and IPA in pmol/10 6 CFU. P-values indicate significant differences between clinical groups determined by one-way ANOVA. (B) Point-biserial correlation heatmap between individual F. nucleatum isolates and specific indole derivatives. The y-axis is color-coded by clinical origin: green (healthy colon), blue (Crohn’s), red (CRC), and orange (ATCC 25586). Asterisks denote statistically significant correlations (* p < 0.05; ** p < 0.01). (C) Ratios of indole derivatives to parent indole, indicating metabolic preference for specific pathways. Each cell displays the ratio value, with colors representing low (blue) to high (red) ratios. Isolates are grouped by clinical association: CRC (n=6), Crohn’s (n=6), healthy colon (n=3), and oral reference strain (ATCC 25586; n=1). F. nucleatum isolates demonstrate variability in biofilm formation across clinical origins Prior to evaluating the effects of exogenous indole treatments, we first quantified baseline biofilm formation over 72 hours to determine if clinical origin inherently influenced F. nucleatum biomass accumulation. Biofilm development was monitored at 12-hour intervals over a 72-hour period. At the 12-hour mark, most isolates exhibited minimal biofilm development, except for the CRC-derived isolate CC2_3FMU1, which displayed the highest early-stage biomass ( Figure 3A ) . By 72 hours, distinct phenotypes emerged; the most robust biofilm formers were predominantly Crohn’s disease-derived (e.g., 7-1 (EAVG002) and EAVG_30), while most CRC-derived isolates, with the notable exception of SB-CTX3Tcol3, produced the lowest baseline biomass ( Figure 3A-B ; Supplementary Figure 4) . Consistent with our indole observations, these phenotypic distinctions diminished when isolates were grouped by subspecies ( Figure 3C ). Collectively, these data highlight substantial differences in biofilm formation across clinical isolates of F. nucleatum . Download figure Open in new tab Figure 3: Variability in biofilm formation among clinical F. nucleatum isolates across different clinical origins. (A) Biofilm formation was quantified at six time points (12, 24, 36, 48, 60, and 72 hours) for 16 clinical isolates. Measurements were normalized to the maximum biofilm value at each time point. Statistical significance for differences between individual isolates was analyzed by adjusted ANOVA (Benjamini–Hochberg correction), followed by Tukey’s HSD test for pairwise comparisons. Significance is denoted as p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****). (B, C) Boxplots displaying normalized biofilm formation of isolates categorized by clinical origin (B) and subspecies (C) . Values are shown as the proportion relative to the maximum biofilm measurement at each time point. Note: Categorization by clinical origin includes CRC (n=6), Crohn’s (n=6), and healthy colon (n=3). The oral isolate (n=1) is included as a single reference point and was excluded from group-based statistical comparisons. F. nucleatum exhibits isolate- and derivative-specific phenotypic shifts in response to exogenous indoles. To ensure that subsequent virulence assays were not confounded by growth inhibition, we evaluated the impact of indole derivative exposure (0.25–2 mM) on isolate growth kinetics using area under the curve (AUC) analysis (Supplementary Figure 6) . While most isolates maintained stable growth up to 1 mM, some strains exhibited significant growth inhibition (e.g., CC53) or stimulatory phenotypes (e.g., CC2_FMU1). Similarly, indole derivatives modulated biofilm formation in a strain-specific manner (Supplementary Figure 7) . To identify which F. nucleatum isolates were most sensitive to indole derivatives, we combined our growth and biofilm findings (Supplementary Figure 5 and 6) into a single metric: the maximum mean signed deviation ( Figure 4 ) . By calculating the largest deviation from the untreated baseline for both growth and biofilm, we determined the magnitude and direction of each isolate’s general response to that indole. Hierarchical clustering of the max signed deviation distinguished specific response patterns among the isolates ( Figure 4A , Supplementary Figure 12) . Notably, the CRC-derived isolate CC53 emerged as the most universally susceptible strain, exhibiting significant dual repression of both growth and biofilm across nearly all tested derivatives (p<0.001). In contrast, other sensitive strains like CC2_6JVN3 and ATCC 25586 displayed a decoupled stress response, characterized by severe growth attenuation paired with biofilm induction (p<0.05). Download figure Open in new tab Figure 4. Fusobacterium nucleatum exhibits isolate- and derivative-specific phenotypic shifts in response to exogenous indoles. (A) Heatmap depicting the maximum signed deviation (calculated as the maximum mean difference (Δ) between the treatment and the untreated control across all tested concentrations) in relative biofilm mass and overall growth (AUC) for isolates exposed to indole and its derivatives. Phenotypic induction and repression are shown in red and blue, respectively. Isolates are hierarchically clustered (Ward’s method, Euclidean distance) and color-coded by clinical origin (red: CRC; blue: Crohn’s; green: Healthy colon; orange: Oral Reference). Significance versus untreated controls (* p < 0.05, ** p < 0.01, *** p < 0.001) was determined via one-way ANOVA with Benjamini-Hochberg FDR correction. (B) Bivariate scatter plots of coupled biofilm and growth maximum signed deviation responses per treatment. The origin (0,0) represents the untreated baseline. Point size and color denote the overall Response Magnitude, defined as the Euclidean distance from the baseline (√Δ Growth 2 + Δ Biofilm 2 ), highlighting hyper-responsive isolates in yellow. To define which isolates were hyper-responsive, we mapped these coupled growth and biofilm changes and calculated each strain’s total response magnitude ( Figure 4B ) . This bivariate map separated the hyper-responsive isolates (CC53, EAVG_027, CC2_6JVN3, and 7-1 (EAVG002)), which consistently fell outside the 75% population confidence interval, the boundary representing the general response observed across the isolates. In contrast, indole-insensitive isolates clustered tightly around the origin (0,0), demonstrating their broad tolerance to indole-mediated stress across both growth and biofilm phenotypes. Interestingly, among hyper-responsive isolates, the nature of the sensitivity was highly specific to both the isolate and the specific indole derivative. While CC2_6JVN3 was universally responsive to every indole tested, other strains exhibited much more selective sensitivities; for instance, CC53 was predominantly affected by I3A, while 7-1 (EAVG002) was most sensitive to I3CA and IPA. These findings suggest that while most F. nucleatum isolates remain insensitive to exogenous indole derivatives, specific clinical isolates possess distinct vulnerabilities that are triggered by specific indole derivatives. Endogenous indole production is not a primary predictor of sensitivity to exogenous indole derivatives We next asked whether an isolate’s endogenous indole production could predict its sensitivity to exogenous treatment ( Figure 5 ). Comparison of total indole (Kovac’s reagent) with phenotypic responses revealed no significant correlation in either biofilm mass (ρ = 0.26, p = 0.338) or growth (ρ = 0.31, p = 0.239) ( Figure 5A ) . In fact, the isolates that produced the most indole (SB-CTX47T, SB-CTX3Tcol3) were largely insensitive to indole treatment except for the isolate CC7_3JVN3C1. The most sensitive isolates (EAVG_012, EAVG_027, 7-1 (EAVG002), CC53, EAVG005, and CC2_6JVN3) were low to moderate indole producers. To achieve higher resolution, we used HPLC to correlate the endogenous production of specific derivatives with their corresponding exogenous effects ( Figure 5B ) . Consistent with the Kovacs data, most indole derivatives exhibited a decoupled relationship between production and sensitivity. A notable exception occurred with I3CA, where endogenous levels showed a moderate negative correlation with biofilm response (ρ = −0.51, p = 0.048). This trend was largely driven by a subset of isolates (including 7-1 (EAVG002), CC7_3JVN3C1, CC53, and SB-CTX3Tcol3), which were both high I3CA producers and highly susceptible to I3CA-mediated biofilm inhibition. Together, these results suggest that F. nucleatum sensitivity to exogenous indoles are largely independent of its own indole production. Download figure Open in new tab Figure 5: Endogenous indole production is independent of sensitivity to exogenous indoles. (A) Correlation between maximum endogenous indole production (quantified via Kovacs assay) and peak phenotypic response to exogenous indole (Max Signed Δ) across F. nucleatum isolates. Point fill scale corresponds to indole production concentration (µM). (B) Correlation between the endogenous production of specific indole derivatives (quantified via HPLC, log 10 pmol per 10 6 CFU) and response to the corresponding exogenous treatment. For all plots, dashed lines represent linear regression trends with 95% confidence intervals (gray shading). Spearman correlation coefficients ( rho = ρ ), R 2 values, and p -values are displayed. Transcriptional profiling reveals strain-specific heterogeneity and directional expression signatures To determine if indole exposure induces distinct transcriptional responses, we quantified the expression of four virulence-associated genes ( aid1 , fomA , fadA , radD ) and the tryptophanase gene ( tnaA ) across the 16 isolates following 72-hour exposure. Mapping these profiles onto a maximum likelihood cladogram revealed heterogeneous responses independent of genetic relatedness ( Figure 6A ) . Closely related isolates exhibited divergent expression (e.g., vincentii vs. CC53), while distantly related isolates showed convergent fomA upregulation. To evaluate whether these expression patterns correlate with clinical origin, we utilized a Naïve Bayes classification model. Certain genes, notably fomA , showed disease-specific expression signatures. These distinct frequency patterns ( Figure 6B ) provided the conditional probability framework for our Naïve Bayes classification model. The categorical expression direction model (Up vs. Down) yielded an AUC of 0.77 ( accuracy = 0.58, p = 0.026 ), outperforming raw fold-change values ( Figure 6C , Supplementary Figure 13B) . Precision-recall analysis further supported the stability of the directional model across cohorts ( Figure 6E , Supplementary Figure 14) . Feature importance and univariate analyses identified fomA , tnaA , and aid1 as the primary drivers of this association ( Figure 6D , Supplementary Figure 13D) . While these data suggest a preliminary association between indole-induced transcriptional shifts and disease origin, we acknowledge the high risk of model overfitting given the limited sample size (n=16); consequently, these patterns represent exploratory, isolate-specific traits rather than definitive diagnostic markers. Moreover, clinical labels were scattered throughout the cladogram ( Figure 6A ) , showing that neither taxonomy nor disease origin dictates a uniform response. This disconnect suggests that indole sensitivity may be unique trait of individual isolates, rather than a shared evolutionary characteristic. Download figure Open in new tab Figure 6. Transcriptional profiling reveals strain-specific heterogeneity and directional expression signatures. (A) 16S rRNA-based maximum likelihood cladogram of 16 F. nucleatum isolates paired with a heatmap showing expression of targeted virulence genes in response to 1 mM indole (72 h). Heatmap displays relative fold expression (2 - ΔΔ Ct , normalized to 16S rRNA and vehicle control). Expression direction threshold is categorized as Up (fold > 1.25; red), Down (fold < 0.75; blue), or No Change (white; gray = undetected). Significance was determined via pairwise t-tests with Bonferroni correction (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). (B) Conditional probabilities ( P (Direction | Clinical Origin) ) of gene expression states across clinical cohorts. These frequencies were derived from the heatmap in (A) and served as training features for the Naïve Bayes classifier. (C) Receiver Operating Characteristic (ROC) curve assessing Naïve Bayes performance (true positive vs. false positive rate) using categorical Expression Direction versus quantitative Fold Change. (D) Feature importance analysis of the categorical expression model. Scores show the AUC for individual genes, reflecting their independent capacity to classify a specific disease state against all others (0.5 = random prediction). (E) Precision-recall curves evaluating the trade-off between precision (positive predictive value) and recall (sensitivity) when classifying disease origin with the categorical versus quantitative models. Indole derivatives strengthen epithelial tight junction integrity and restrict bacterial invasion in CRC cell models To determine if indole derivatives could influence virulence independently of growth and biofilm dynamics, we selected isolate SB-CTX3Tcol3 as this isolate remains phenotypically insensitive to treatment despite its high levels of endogenous indole production and biofilm formation. A bacterial invasion assay using Caco-2 cells demonstrated that indole treatment for 2 or 4 hours markedly inhibited invasion, producing a pathogen-suppressive effect comparable to the presence of antibiotics in the medium ( Figure 7A ). Notably, I3CA exhibited the strongest early inhibitory effect, reducing invasion by approximately 20% after only 2 hours. By 4 hours, treatment with indole, I3A, and I3CA suppressed invasion by more than 50%, whereas IAA and IPA had no detectable impact on the invasion capacity of SB-CTX3Tcol3. These data highlight that indole-induced effects are specific to the indole species present. The reliance of F. nucleatum on adhesins such as FadA to bind E-cadherin and promote invasion prompted us to investigate whether indoles could strengthen the intestinal barrier to prevent this process. When Caco-2 cells were treated with indole derivatives at the concentrations used in the invasion assay, transcripts for tight and adherens junction genes were predominantly upregulated ( Figure 7B ). Notably, I3A exerted the strongest effect on transcript levels, followed by indole, while I3CA, IAA, and IPA showed minimal impact on transcript expression. Similar, though less pronounced, trends were observed in HT-29 cells. These findings suggest that certain indoles enhance intestinal barrier integrity to help prevent pathogen invasion. However, the strong inhibitory effect of I3CA on bacterial entry despite its low impact on host gene transcripts suggests it may operate through a distinct, yet unidentified, mechanism. Download figure Open in new tab Figure 7: Indole derivatives strengthen epithelial tight junction integrity and restrict bacterial invasion in CRC cell models. (A) Invasion of a high-indole, high-biofilm producing CRC isolate (SB-CTX3Tcol3) into Caco-2 cells over 2 and 4 hours. F. nucleatum cultures were adjusted to a multiplicity of infection (MOI) of 50:1 and added to Caco-2 cells treated with 1 mM indole derivatives, vehicle control (0.1% DMSO), or a combination of penicillin (200 U/mL) and streptomycin (200 µg/mL) as a negative control (Anti). Statistical significance was assessed by Wilcoxon signed-rank test: ***p < 0.001, **p < 0.01, *p < 0.05. (B) Caco-2 and HT-29 cells were treated for 24 hours with 1 mM indole derivatives, vehicle control (0.1% DMSO), or lipopolysaccharide (LPS, 10 µg/mL) as a positive control for barrier disruption. Relative fold expression change (treatment vs. DMSO control) for tight junction genes ( CLDN1, CLDN2, CLDN7 ) and adherens junction genes ( OCLN, CDH1 ) was analyzed by qPCR and calculated using the ΔΔCt method. Heatmap colors represent the average fold change across four biological replicates, with pairwise comparisons assessed by Wilcoxon signed-rank test. DISCUSSION In this pilot study, we investigated whether bacteria isolated from disease environments retain niche-specific virulence programs, particularly those governing biofilm formation and indole production. Using F. nucleatum as a model, we profiled indole production and sensitivity across 16 clinical isolates. Our central findings are that F. nucleatum clinical isolates exhibit significant strain-level heterogeneity across multiple phenotypic metrics, with diversity being attributed to individual isolates as opposed to disease origin or subspecies. Specifically, we observed : (i) substantial variation in indole production, with some isolates producing 3–4-fold more indole than average and a subset biased toward I3A derivative production; (ii) a wide range of biofilm formation capacities; (iii) isolate-dependent sensitivity to exogenous indole derivatives, including hyper-responsive clinical isolates; (iv) a relationship between increased I3CA production and I3CA-mediated biofilm inhibition; (v) highly heterogeneous, isolate-specific transcriptional responses to indole derivatives, where the direction of gene expression provides a stronger classification signal than expression magnitude; and (vi) indole derivative–specific strengthening of epithelial tight junctions while simultaneously reducing bacterial invasion. Together, these findings suggest that clinical isolates of F. nucleatum exhibit diverse indole metabolic programs that may influence virulence and adaptation within disease-associated environments. Indole production is a defining feature of gut microbiota ( 37 ); however, the specific metabolic contributions of opportunistic pathogens like F. nucleatum within this network remain poorly defined. While recent evidence implicates tryptophan metabolites in inflammatory conditions and CRC pathogenesis ( 28 , 38 , 39 ), our data demonstrate that F. nucleatum is not a uniform participant in this process. Rather than exhibiting a conserved metabolic profile, we observed profound isolate-level heterogeneity in baseline indole production. Crucially, our isolates produced a diverse array of indole derivatives, most notably I3A. This metabolite was preferentially produced by a subset of CRC-derived isolates, consistent with previous reports identifying I3A as differentially abundant in CRC patient cohorts ( 36 ). Together, these findings demonstrate strain-specific metabolic diversity suggesting a functional consequence within the host. Current understanding of F. nucleatum indole production is limited and disproportionately based on the ATCC 25586 reference strain ( 24 , 30 , 40 ). While IPA production in ATCC 25586 reportedly drives CRC progression via AhR-mediated M2 macrophage polarization ( 30 ), our data reveal that IPA is the least prevalent derivative across the clinical isolates examined, frequently falling below the limit of detection ( Figure 3A-C , Supplementary Figure 3A) . This discrepancy suggests that IPA-driven procarcinogenic pathways may be isolate-specific rather than universal. By characterizing a broader range of clinical isolates, this work demonstrates that a single reference strain cannot capture the metabolic heterogeneity underlying indole’s role in F. nucleatum pathogenesis. These differences extend beyond indole production, as individual isolates also vary in how they respond to indole in their environment. A defining characteristic of our study is the heterogeneity in F. nucleatum sensitivity to exogenous indole derivatives. While most responsive isolates (e.g., CC53) showed coordinated changes with growth and biofilm formation declining together, strains like CC2_6JVN3 and ATCC 25586 displayed a decoupled response, with biofilm formation increasing even as growth slowed. This pattern suggests that, for these isolates, indole may act as an environmental stressor, triggering a shift from planktonic growth to a protective biofilm state, a survival strategy observed in other bacterial pathogens ( 41 ). While most isolates exhibited only minor phenotypic variation, bivariate analysis of signed deviations highlighted three hyper-responsive strains: CC2_6JVN3, CC53, and 7-1 (EAVG002). CC2_6JVN3 showed broad sensitivity across nearly all derivatives, whereas CC53 and 7-1 displayed more selective vulnerabilities (7-1 primarily to I3CA and IPA, and CC53 to I3A). These sensitivities were largely independent of endogenous production, except for I3CA, where higher production correlated with greater biofilm inhibition (ρ = −0.51, p = 0.048). In contrast, robust indole producers like SB-CTX3Tcol3 remained largely insensitive, suggesting that these isolates may have evolved mechanisms to tolerate high local indole concentrations. Notably, the strongest phenotypic effects were observed at 2 mM, which may exceed typical bulk fecal indole concentrations; however, local concentrations within biofilms or the tumor microenvironment remain unknown and could be substantially higher ( 42 ). Together, these findings highlight that F. nucleatum strains display highly strain-specific sensitivity and insensitivity to indole derivatives. Interestingly, the insensitivity to indoles appears to be decoupled from F. nucleatum invasion. At physiologically relevant concentrations ( 35 , 43 ), indole reduced invasion by ∼50% in the otherwise indole-insensitive and highest indole-producing isolate SB-CTX3Tcol3, indicating that indole can modulate invasive behavior independently of its antimicrobial or biofilm effects. These findings suggest that indole acts as a regulatory signal, suppressing active pathogenicity while simultaneously limiting population growth, as reflected in the reduced carrying capacity observed across most isolates (Supplementary Figure 5) . Given the high variability in both endogenous indole production and sensitivity, further mechanistic studies across a broader panel of clinical isolates are needed to definitively clarify indole’s role in F. nucleatum virulence. Given our small sample size (n=16), any classification approach carries substantial overfitting risk. We therefore employed a Naïve Bayes classifier, leveraging its high-bias, low-variance architecture. The model achieved an AUC of 0.77; however, its modest accuracy ( 0.58, p = 0.021 vs. NIR ) suggests it is better suited for exploratory pattern discovery than clinical diagnosis. Specifically, the analysis revealed that the direction of gene expression (up- vs. downregulation) predicted clinical origin more reliably than fold-change magnitude. This indicates that the core transcriptional logic of the adhesin fomA is highly conserved, even when the absolute strength of the response varies. As directional fomA expression was a primary classifier and was consistently upregulated in CRC-derived isolates, it may represent a viable target for strain-selective antimicrobials. Biologically, FomA serves as a critical anchor for host epithelial adherence and has demonstrated immune-adjuvant activity in mammalian models ( 16 , 17 ). This directional signal positions FomA as a compelling candidate for strain-selective therapeutic targeting, aligning with recent work demonstrating its viability as a target for precision antimicrobial peptide therapies ( 44 ). However, the broad dispersion of clinical origins across our cladogram confirms that these responses do not strictly track with phylogeny. Therefore, they should be considered exploratory signatures rather than definitive disease markers. It is essential to acknowledge that this study serves as a pilot investigation. The limited number of isolates examined (n=16), the singular oral reference strain, and the inherent overfitting risks associated with modeling high-dimensional data on small cohorts mean that broader generalizations regarding disease or subspecies specificity require rigorous validation. Future research utilizing dual-RNA-seq during active invasion will be necessary to define the precise host-pathogen molecular pathways by which indoles impede bacterial entry. Despite these limitations, our findings establish that clinical isolates of F. nucleatum exhibit profound strain-level heterogeneity in indole metabolism and sensitivity. This metabolic diversity highlights vulnerabilities that represent candidate targets that can be exploited for precision or strain-selective therapeutic interventions. DISCLOSURE STATEMENT The authors declare that they have no known competing financial interests or personal relationships that could have influenced the work reported in this paper. FUNDING Funding for this project was provided by a grant to Leigh Greathouse from the Baylor ONE-URC Grant Award. CRediT AUTHORSHIP CONTRIBUTION STATEMENT C.S. Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Validation, Visualization, Writing - original draft, Writing - review & editing. A.C. Conceptualization, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing – review & editing. M.R. Investigation, Methodology, Writing – review & editing. J.H. Data curation, Investigation, Writing – review & editing. L.C. Investigation, Writing – review & editing. G.Z. Investigation, Writing – review & editing. R.L. Conceptualization, Resources, Writing – review & editing. L.G. Corresponding author, Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Writing – original draft, Writing – review & editing. DATA AVAILABILITY STATEMENT The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials. All data generated or analyzed during this study have been deposited in the Texas Data Repository ( https://doi.org/10.18738/T8/3PYJA9 ) and are available upon request. Repositories The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials. All data generated or analyzed during this study have been deposited in the Texas Data Repository ( https://doi.org/10.18738/T8/3PYJA9 ) and are available upon request. View this table: View inline View popup Download powerpoint Table 1. Sources and references of clinical Fusobacterium nucleatum isolates. A summary of clinical F. nucleatum isolates obtained from various disease contexts, including healthy colon controls, Crohn’s disease (remission and active), colorectal cancer, and oral lesions. Isolates are categorized by strain, alternative names (based on PubMed identifiers), sampling site, associated disease state, and subspecies classification. Subspecies was determined with subspecies specific primers (Supplemental Table 1). Supplier information and relevant literature references are also provided. ACKNOWLEDGMENTS The authors thank the Molecular Biosciences Core (MBC) at Baylor University for access to instrumentation and expert technical assistance. An earlier pre-print version of this manuscript is available on bioRxiv ( 45 ). We would also like to thank Dr. Emma Allen Vercoe and Dr. Susan Bullman for gifting the clinical F. nucleatum isolates. Footnotes New analysis, new figures, and improved writing. REFERENCES 1. ↵ Kostic AD , Gevers D , Pedamallu CS , Michaud M , Duke F , Earl AM , et al. Genomic analysis identifies association of Fusobacterium with colorectal carcinoma . Genome Res . 2012 Feb ; 22 ( 2 ): 292 – 8 . doi: 10.1101/gr.126573.111 PubMed PMID: 22009990 ; PubMed Central PMCID: PMC3266036 . OpenUrl Abstract / FREE Full Text 2. ↵ Castellarin M , Warren RL , Freeman JD , Dreolini L , Krzywinski M , Strauss J , et al. Fusobacterium nucleatum infection is prevalent in human colorectal carcinoma . Genome Res . 2012 Feb ; 22 ( 2 ): 299 – 306 . doi: 10.1101/gr.126516.111 PubMed PMID: 22009989 ; PubMed Central PMCID: PMC3266037 . OpenUrl Abstract / FREE Full Text 3. ↵ Abed J , Maalouf N , Manson AL , Earl AM , Parhi L , Emgård JEM , et al. Colon Cancer-Associated Fusobacterium nucleatum May Originate From the Oral Cavity and Reach Colon Tumors via the Circulatory System . Front Cell Infect Microbiol [Internet] . 2020 [cited 2022 Nov 30 ]; 10 . Available from: https://www.frontiersin.org/articles/10.3389/fcimb.2020.00400 4. ↵ Ma X , Sun T , Zhou J , Zhi M , Shen S , Wang Y , et al. Pangenomic Study of Fusobacterium nucleatum Reveals the Distribution of Pathogenic Genes and Functional Clusters at the Subspecies and Strain Levels . Microbiol Spectr . 2023 Jun 15 ; 11 ( 3 ): e0518422 . doi: 10.1128/spectrum.05184-22 PubMed PMID: 37042769 ; PubMed Central PMCID: PMC10269558 . OpenUrl CrossRef PubMed 5. ↵ Queen J , Cing Z , Minsky H , Nandi A , Southward T , Ferri J , et al. Fusobacterium nucleatum is enriched in invasive biofilms in colorectal cancer . NPJ Biofilms Microbiomes . 2025 May 20 ; 11 : 81 . doi: 10.1038/s41522-025-00717-7 PubMed PMID: 40394001 ; PubMed Central PMCID: PMC12092649 . OpenUrl CrossRef PubMed 6. ↵ Cao P , Chen Y , Guo X , Chen Y , Su W , Zhan N , et al. Fusobacterium nucleatum Activates Endoplasmic Reticulum Stress to Promote Crohn’s Disease Development via the Upregulation of CARD3 Expression . Front Pharmacol . 2020 Feb 21 ; 11 : 106 . doi: 10.3389/fphar.2020.00106 PubMed PMID: 32153411 ; PubMed Central PMCID: PMC7047714 . OpenUrl CrossRef PubMed 7. ↵ Strauss J , Kaplan GG , Beck PL , Rioux K , Panaccione R , DeVinney R , et al. Invasive potential of gut mucosa-derived fusobacterium nucleatum positively correlates with IBD status of the host . Inflamm Bowel Dis . 2011 Sep 1 ; 17 ( 9 ): 1971 – 8 . doi: 10.1002/ibd.21606 OpenUrl CrossRef PubMed Web of Science 8. ↵ Tomkovich S , Dejea CM , Winglee K , Drewes JL , Chung L , Housseau F , et al. Human colon mucosal biofilms from healthy or colon cancer hosts are carcinogenic . J Clin Invest . 129 ( 4 ): 1699 – 712 . doi: 10.1172/JCI124196 PubMed PMID: 30855275 ; PubMed Central PMCID: PMC6436866 . OpenUrl CrossRef PubMed 9. ↵ Dejea CM , Wick EC , Hechenbleikner EM , White JR , Mark Welch JL , Rossetti BJ , et al. Microbiota organization is a distinct feature of proximal colorectal cancers . Proc Natl Acad Sci U S A . 2014 Dec 23 ; 111 ( 51 ): 18321 – 6 . doi: 10.1073/pnas.1406199111 PubMed PMID: 25489084 ; PubMed Central PMCID: PMC4280621 . OpenUrl Abstract / FREE Full Text 10. ↵ Kostic AD , Chun E , Robertson L , Glickman JN , Gallini CA , Michaud M , et al. Fusobacterium nucleatum potentiates intestinal tumorigenesis and modulates the tumor immune microenvironment . Cell Host Microbe . 2013 Aug 14 ; 14 ( 2 ): 207 – 15 . doi: 10.1016/j.chom.2013.07.007 PubMed PMID: 23954159 ; PubMed Central PMCID: PMC3772512 . OpenUrl CrossRef PubMed Web of Science 11. ↵ Zepeda-Rivera M , Minot SS , Bouzek H , Wu H , Blanco-Míguez A , Manghi P , et al. A distinct Fusobacterium nucleatum clade dominates the colorectal cancer niche . Nature . 2024 Apr ; 628 ( 8007 ): 424 – 32 . doi: 10.1038/s41586-024-07182-w OpenUrl CrossRef 12. ↵ Umana A , Sanders BE , Yoo CC , Casasanta MA , Udayasuryan B , Verbridge SS , et al. Reevaluating the Fusobacterium Virulence Factor Landscape [Internet] . bioRxiv ; 2019 [cited 2024 Oct 17 ]. p. 534297 . Available from: https://www.biorxiv.org/content/10.1101/534297v1 doi: 10.1101/534297 OpenUrl Abstract / FREE Full Text 13. Kaplan A , Kaplan CW , He X , McHardy I , Shi W , Lux R . Characterization of aid1, a novel gene involved in Fusobacterium nucleatum interspecies interactions . Microb Ecol . 2014 Aug ; 68 ( 2 ): 379 – 87 . doi: 10.1007/s00248-014-0400-y PubMed PMID: 24643713 ; PubMed Central PMCID: PMC4104215 . OpenUrl CrossRef PubMed 14. Zhang L , Leng XX , Qi J , Wang N , Han JX , Tao ZH , et al. The adhesin RadD enhances Fusobacterium nucleatum tumour colonization and colorectal carcinogenesis . Nat Microbiol . 2024 Sep ; 9 ( 9 ): 2292 – 307 . doi: 10.1038/s41564-024-01784-w OpenUrl CrossRef PubMed 15. ↵ Rubinstein MR , Wang X , Liu W , Hao Y , Cai G , Han YW . Fusobacterium nucleatum promotes colorectal carcinogenesis by modulating E-cadherin/β-catenin signaling via its FadA adhesin . Cell Host Microbe . 2013 Aug 14 ; 14 ( 2 ): 195 – 206 . doi: 10.1016/j.chom.2013.07.012 PubMed PMID: 23954158 ; PubMed Central PMCID: PMC3770529 . OpenUrl CrossRef PubMed Web of Science 16. ↵ Nakagaki H , Sekine S , Terao Y , Toe M , Tanaka M , Ito HO , et al. Fusobacterium nucleatum Envelope Protein FomA Is Immunogenic and Binds to the Salivary Statherin-Derived Peptide . Infect Immun . 2010 Mar ; 78 ( 3 ): 1185 – 92 . doi: 10.1128/IAI.01224-09 OpenUrl Abstract / FREE Full Text 17. ↵ Zhang Z , Liu S , Zhang S , Li Y , Shi X , Liu D , et al. Porphyromonas gingivalis outer membrane vesicles inhibit the invasion of Fusobacterium nucleatum into oral epithelial cells by downregulating FadA and FomA . J Periodontol . 2022 Apr ; 93 ( 4 ): 515 – 25 . doi: 10.1002/JPER.21-0144 PubMed PMID: 34458990 ; PubMed Central PMCID: PMC9415117 . OpenUrl CrossRef PubMed 18. ↵ Xu M , Yamada M , Li M , Liu H , Chen SG , Han YW . FadA from Fusobacterium nucleatum Utilizes both Secreted and Nonsecreted Forms for Functional Oligomerization for Attachment and Invasion of Host Cells . J Biol Chem . 2007 Aug ; 282 ( 34 ): 25000 – 9 . doi: 10.1074/jbc.M611567200 OpenUrl Abstract / FREE Full Text 19. ↵ Guo P , Tian Z , Kong X , Yang L , Shan X , Dong B , et al. FadA promotes DNA damage and progression of Fusobacterium nucleatum-induced colorectal cancer through up-regulation of chk2 . J Exp Clin Cancer Res CR . 2020 Sep 29 ; 39 : 202 . doi: 10.1186/s13046-020-01677-w PubMed PMID: 32993749 ; PubMed Central PMCID: PMC7523382 . OpenUrl CrossRef PubMed 20. ↵ Li D , Li Z , Wang L , Zhang Y , Ning S . Oral inoculation of Fusobacterium nucleatum exacerbates ulcerative colitis via the secretion of virulence adhesin FadA . Virulence . 15 ( 1 ): 2399217 . doi: 10.1080/21505594.2024.2399217 PubMed PMID: 39221673 ; PubMed Central PMCID: PMC11385161 . OpenUrl CrossRef PubMed 21. ↵ Liu PF , Shi W , Zhu W , Smith JW , Hsieh SL , Gallo RL , et al. Vaccination targeting surface FomA of Fusobacterium nucleatum against bacterial co-aggregation: implication for treatment of periodontal infection and halitosis . Vaccine . 2010 Apr 26 ; 28 ( 19 ): 3496 – 505 . doi: 10.1016/j.vaccine.2010.02.047 PubMed PMID: 20189489 ; PubMed Central PMCID: PMC2855893 . OpenUrl CrossRef PubMed 22. ↵ Gursoy UK , Pöllänen M , Könönen E , Uitto VJ . Biofilm Formation Enhances the Oxygen Tolerance and Invasiveness of Fusobacterium nucleatum in an Oral Mucosa Culture Model . J Periodontol . 2010 ; 81 ( 7 ): 1084 – 91 . doi: 10.1902/jop.2010.090664 OpenUrl CrossRef PubMed 23. ↵ Lee JH , Wood TK , Lee J . Roles of Indole as an Interspecies and Interkingdom Signaling Molecule . Trends Microbiol . 2015 Nov 1 ; 23 ( 11 ): 707 – 18 . doi: 10.1016/j.tim.2015.08.001 OpenUrl CrossRef PubMed 24. ↵ Sasaki-Imamura T , Yano A , Yoshida Y . Production of Indole from l-Tryptophan and Effects of These Compounds on Biofilm Formation by Fusobacterium nucleatum ATCC 25586 . Appl Environ Microbiol . 2010 Jul ; 76 ( 13 ): 4260 – 8 . doi: 10.1128/AEM.00166-10 PubMed PMID: 20472741 ; PubMed Central PMCID: PMC2897440 . OpenUrl Abstract / FREE Full Text 25. ↵ Feng Q , Liang S , Jia H , Stadlmayr A , Tang L , Lan Z , et al. Gut microbiome development along the colorectal adenoma–carcinoma sequence . Nat Commun . 2015 Mar 11 ; 6 ( 1 ): 1 . doi: 10.1038/ncomms7528 OpenUrl CrossRef PubMed 26. Zeller G , Tap J , Voigt AY , Sunagawa S , Kultima JR , Costea PI , et al. Potential of fecal microbiota for early-stage detection of colorectal cancer . Mol Syst Biol . 2014 Nov 28 ; 10 ( 11 ): 766 . doi: 10.15252/msb.20145645 PubMed PMID: 25432777 ; PubMed Central PMCID: PMC4299606 . OpenUrl Abstract / FREE Full Text 27. ↵ Vogtmann E , Hua X , Zeller G , Sunagawa S , Voigt AY , Hercog R , et al. Colorectal Cancer and the Human Gut Microbiome: Reproducibility with Whole-Genome Shotgun Sequencing . PLoS ONE . 2016 May 12 ; 11 ( 5 ): e0155362 . doi: 10.1371/journal.pone.0155362 PubMed PMID: 27171425 ; PubMed Central PMCID: PMC4865240 . OpenUrl CrossRef PubMed 28. ↵ Priyadarshini M , Jorgensen J , Stauffer SRC , Issa L , Pandya N , Nnyamah C , et al. A high fermentable fiber Western diet reduces indole levels . BioRxiv Prepr Serv Biol . 2026 Jan 30 ; 2026.01.27.702025 . doi: 10.64898/2026.01.27.702025 PubMed PMID: 41659626 ; PubMed Central PMCID: PMC12873831 . OpenUrl Abstract / FREE Full Text 29. ↵ Won KJ , Jin PR , Davis L , Orjala J , Armstrong AT , Wang WH , et al. Fusobacterium nucleatum produces previously unappreciated AHR-activating metabolites and promotes CRC cell proliferation via the AHR-TERT axis . BioRxiv Prepr Serv Biol . 2025 Sep 25 ; 2025.09.25.678601 . doi: 10.1101/2025.09.25.678601 PubMed PMID: 41040148 ; PubMed Central PMCID: PMC12485871 . OpenUrl Abstract / FREE Full Text 30. ↵ Song Q , Jin Z , Zhang H , Hong K , Zhu B , Yin H , et al. Fusobacterium nucleatum -derived 3-indolepropionic acid promotes colorectal cancer progression via aryl hydrocarbon receptor activation in macrophages . Chem Biol Interact . 2025 Jun 1 ; 414 : 111495 . doi: 10.1016/j.cbi.2025.111495 OpenUrl CrossRef 31. ↵ Tuomola M , Vahva M , Kallio H . High-Performance Liquid Chromatography Determination of Skatole and Indole Levels in Pig Serum, Subcutaneous Fat, and Submaxillary Salivary Glands . J Agric Food Chem . 1996 Jan 1 ; 44 ( 5 ): 1265 – 70 . doi: 10.1021/jf950796z OpenUrl CrossRef 32. ↵ Kema IP , Meijer WG , Meiborg G , Ooms B , Willemse PHB , De Vries EGE . Profiling of Tryptophan-related Plasma Indoles in Patients with Carcinoid Tumors by Automated, On-Line, Solid-Phase Extraction and HPLC with Fluorescence Detection . Clin Chem . 2001 Oct 1 ; 47 ( 10 ): 1811 – 20 . doi: 10.1093/clinchem/47.10.1811 OpenUrl Abstract / FREE Full Text 33. ↵ Manson McGuire A , Cochrane K , Griggs AD , Haas BJ , Abeel T , Zeng Q , et al. Evolution of Invasion in a Diverse Set of Fusobacterium Species . mBio . 2014 Nov 4 ; 5 ( 6 ): e01864 – 14 . doi: 10.1128/mBio.01864-14 OpenUrl CrossRef PubMed 34. ↵ Strauss J , White A , Ambrose C , McDonald J , Allen-Vercoe E . Phenotypic and genotypic analyses of clinical Fusobacterium nucleatum and Fusobacterium periodonticum isolates from the human gut . Anaerobe . 2008 Dec 1 ; 14 ( 6 ): 301 – 9 . doi: 10.1016/j.anaerobe.2008.12.003 OpenUrl CrossRef PubMed 35. ↵ Darkoh C , Chappell C , Gonzales C , Okhuysen P . A rapid and specific method for the detection of indole in complex biological samples . Appl Environ Microbiol . 2015 Dec ; 81 ( 23 ): 8093 – 7 . doi: 10.1128/AEM.02787-15 PubMed PMID: 26386049 ; PubMed Central PMCID: PMC4651089 . OpenUrl Abstract / FREE Full Text 36. ↵ Sun XZ , Zhao DY , Zhou YC , Wang QQ , Qin G , Yao SK . Alteration of fecal tryptophan metabolism correlates with shifted microbiota and may be involved in pathogenesis of colorectal cancer . World J Gastroenterol . 2020 Dec 7 ; 26 ( 45 ): 7173 – 90 . doi: 10.3748/wjg.v26.i45.7173 PubMed PMID: 33362375 ; PubMed Central PMCID: PMC7723673 . OpenUrl CrossRef PubMed 37. ↵ Wikoff WR , Anfora AT , Liu J , Schultz PG , Lesley SA , Peters EC , et al. Metabolomics analysis reveals large effects of gut microflora on mammalian blood metabolites . Proc Natl Acad Sci U S A . 2009 Mar 10 ; 106 ( 10 ): 3698 – 703 . doi: 10.1073/pnas.0812874106 PubMed PMID: 19234110 ; PubMed Central PMCID: PMC2656143 . OpenUrl Abstract / FREE Full Text 38. ↵ Hulin A , Rifflet A , Castelli F , Giai Gianetto Q , Fenaille F , Aissat A , et al. Potential Impact of Microbial Dysbiosis and Tryptophan Metabolites in Advanced Stages of Colorectal Cancer . Biomedicines . 2026 Jan ; 14 ( 1 ): 26 . doi: 10.3390/biomedicines14010026 OpenUrl CrossRef 39. ↵ Zhang C , Rao Z , Zhan X , Qin J , Yang L , Yin Q , et al. Deciphering tryptophan metabolism in colorectal cancer through multi-omics analysis . Biochem Biophys Rep . 2025 Aug 12 ; 43 : 102157 . doi: 10.1016/j.bbrep.2025.102157 PubMed PMID: 40836965 ; PubMed Central PMCID: PMC12361607 . OpenUrl CrossRef PubMed 40. ↵ Xue Y , Xiao H , Guo S , Xu B , Liao Y , Wu Y , et al. Indoleamine 2,3-dioxygenase expression regulates the survival and proliferation of Fusobacterium nucleatum in THP-1-derived macrophages . Cell Death Dis . 2018 Mar 2 ; 9 ( 3 ): 355 . doi: 10.1038/s41419-018-0389-0 OpenUrl CrossRef PubMed 41. ↵ Fang X , Chen Y , Hu L , Gu S , Zhu J , Hang Y , et al. Modulation of virulence and metabolic profiles in Klebsiella pneumoniae under indole-mediated stress response . Front Cell Infect Microbiol . 2025 ; 15 : 1546991 . doi: 10.3389/fcimb.2025.1546991 PubMed PMID: 39963410 ; PubMed Central PMCID: PMC11830697 . OpenUrl CrossRef PubMed 42. ↵ Shimizu H , Niwa T . Indoxyl Sulfate in the Gut–Kidney Axis: Pathophysiology and Clinical Significance in CKD-Associated Colorectal Cancer . Toxins . 2026 Feb ; 18 ( 2 ): 72 . doi: 10.3390/toxins18020072 OpenUrl CrossRef PubMed 43. ↵ Gaimster H , Summers D . Regulation of Indole Signalling during the Transition of E. coli from Exponential to Stationary Phase . PLoS ONE . 2015 Sep 2 ; 10 ( 9 ): e0136691 . doi: 10.1371/journal.pone.0136691 PubMed PMID: 26332864 ; PubMed Central PMCID: PMC4557909 . OpenUrl CrossRef PubMed 44. ↵ Choudhury A , Scano C , Barton A , Kearney CM , Greathouse KL . Precision Antimicrobial Therapy Against Fusobacterium nucleatum Using Bioengineered Probiotics Expressing Guided Antimicrobial Peptides (gAMPs) . Microb Biotechnol . 2025 Dec 10 ; 18 ( 12 ): e70241 . doi: 10.1111/1751-7915.70241 PubMed PMID: 41368940 ; PubMed Central PMCID: PMC12690405 . OpenUrl CrossRef PubMed 45. ↵ Scano C , Choudhury A , Rojo MG , Lavado R , Zaharas G , Hawkins J , et al. Comparative Analysis of Indole Derivative Production Among Clinical Isolates of Fusobacterium nucleatum [Internet] . bioRxiv ; 2025 [cited 2025 Jul 17 ]. p. 2025.05.16.654558 . Available from: https://www.biorxiv.org/content/10.1101/2025.05.16.654558v3 doi: 10.1101/2025.05.16.654558 OpenUrl Abstract / FREE Full Text View the discussion thread. Back to top Previous Next Posted May 11, 2026. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Clinical isolates of Fusobacterium nucleatum display strain-specific virulence and modulation by indole derivatives Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Clinical isolates of Fusobacterium nucleatum display strain-specific virulence and modulation by indole derivatives Colin Scano , Ankan Choudhury , Macarena G. Rojo , Jessalyn Hawkins , Lauren Crowhurst , Greg Zaharas , Ramon Lavado , Leigh Greathouse bioRxiv 2025.05.16.654558; doi: https://doi.org/10.1101/2025.05.16.654558 Share This Article: Copy Citation Tools Clinical isolates of Fusobacterium nucleatum display strain-specific virulence and modulation by indole derivatives Colin Scano , Ankan Choudhury , Macarena G. Rojo , Jessalyn Hawkins , Lauren Crowhurst , Greg Zaharas , Ramon Lavado , Leigh Greathouse bioRxiv 2025.05.16.654558; doi: https://doi.org/10.1101/2025.05.16.654558 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Microbiology Subject Areas All Articles Animal Behavior and Cognition (7642) Biochemistry (17715) Bioengineering (13907) Bioinformatics (42005) Biophysics (21472) Cancer Biology (18624) Cell Biology (25534) Clinical Trials (138) Developmental Biology (13390) Ecology (19935) Epidemiology (2067) Evolutionary Biology (24356) Genetics (15617) Genomics (22529) Immunology (17753) Microbiology (40437) Molecular Biology (17200) Neuroscience (88697) Paleontology (667) Pathology (2840) Pharmacology and Toxicology (4829) Physiology (7653) Plant Biology (15171) Scientific Communication and Education (2046) Synthetic Biology (4304) Systems Biology (9827) Zoology (2272)