Effect of the gut microbiota-derived tryptophan metabolite indole-3-acetic acid in pneumonia

Abstract

Gut microbiota influence the severity of pneumonia by producing metabolites that enhance systemic and pulmonary immune responses. Preclinical studies suggested that gut microbiota-derived indoles have protective effects against numerous diseases, including influenza and abdominal infections. However, the precise role of tryptophan metabolites during pneumonia is unknown. Here, we perform translational analyses in a large general-population cohort (n = 13,464), critically ill patients with severe community-acquired pneumonia (CAP; n = 158; NCT01905033), a randomized human intervention trial on antibiotic-mediated microbiota modulation (NCT03051698), and mice to investigate the effects of tryptophan metabolites, specifically indole-3-acetic acid (IAA), on pneumonia. In the population-based cohort, baseline IAA is associated with a higher risk of future hospital admission for pneumonia (cause-specific hazard ratio 1.15, 95% confidence interval 1.09-1.22 p < 0.0001). In patients with severe CAP higher levels of IAA are associated with increased mortality, independent from potential confounders (hazard ratio 1.30 per log2 increase, 95% confidence interval 1.02-1.68, p = 0.037). In a mouse model of bacterial pneumonia, IAA supplementation aggravates pulmonary damage while reducing systemic dissemination, which is mediated by the aryl hydrocarbon receptor (AhR) and increased release of reactive oxygen species from neutrophils. In summary, these findings from general population and severe pneumonia cohorts, and murine pneumonia experiments, show that the gut microbiota-derived tryptophan metabolite IAA affects pneumonia, suggesting that various indoles may have diverging, context-dependent effects.

The gut microbiota can influence the severity of pneumonia through the production of metabolites. In this translational study, the authors investigate the effects of tryptophan metabolites, specifically indole-3-acetic acid (IAA), on pneumonia.

Introduction

Pneumonia, an acute infection of the lower respiratory tract, affects over 450 million people and causes 2.5 million deaths each year¹. Yet, disease severity varies broadly: the majority of community-acquired pneumonia (CAP) patients can be treated on an outpatient basis, whereas some require hospital or even intensive care unit (ICU) admission^2,3. Knowledge on the biological processes contributing to this heterogeneity in disease severity may advance the prevention and treatment of CAP.

Our group and others identified the intestinal microbiota as modulator of pneumonia severity^4–6. Depletion of gut microbiota by broad-spectrum antibiotics increased the severity of experimental pneumonia in mice^4,6–8. In patients hospitalized for CAP, intestinal microbiota are disrupted and correlate with disease severity^9–11. Also prior to hospitalization, gut microbiota alterations may contribute to heterogeneity in infection severity: we found that depletion of anaerobic gut bacteria was associated with an increased risk of hospital admission in two general population cohorts with 10,699 participants¹². Intestinal communities produce metabolites that translocate from the gut, via the bloodstream, to the lungs and enhance systemic and pulmonary immune responses—a process termed the gut-lung axis^13,14. Virtually all studies investigating this protective gut-lung axis focused on the effects of short-chain fatty acids. Yet, other microbiota-derived metabolites, such as metabolites of the amino acid tryptophan, may contribute to the effects of gut microbiota on pneumonia severity.

In the gut, dietary tryptophan can be metabolized into three major pathways: the serotonin, kynurenine and indole pathway. While the serotonin and kynurenine pathways are largely mediated by host cells, it has been reported that intestinal bacteria transform tryptophan into indoles. Indoles bind to and active the nuclear aryl hydrocarbon receptor (AhR) to elicit downstream gene expression^15,16. Indole metabolite functions vary and have yet to be fully characterized, but certain indole derivates have been linked with enhanced intestinal barrier function, enriched T helper 17 cell differentiation, and increased interleukin (IL)−22 production in the gut, resulting in protection against colitis^15–18. Moreover, specific microbiota-derived indoles, such as indole-3-acetic acid (IAA) and indole-3-propionic acid (IPA), and AhR signaling may also protect against diseases outside of the gut^19–27. For example, IPA improved survival in experimental polymicrobial abdominal sepsis, and reduced viral load and pulmonary inflammation during influenza^25,26, suggesting that IPA may limit infection severity. In addition, AhR signaling enhanced lung barrier function and reduced vessel leakage during influenza-induced pneumonia²⁷. IAA was enriched in pancreatic cancer patients who respond to chemotherapy, and IAA administration enhanced the efficacy of chemotherapy in mice by increasing reactive oxygen species (ROS) release from neutrophils, resulting in tumor cell death²¹. Furthermore, IAA was reduced in sputum of patients with chronic obstructive pulmonary disease (COPD), and protected against COPD and allergic asthma in mice^22,23, which both are important risk factors for pneumonia^2,24. These protective effects prompted calls for clinical trials investigating the effects of targeting IAA in (pancreatic) cancer and COPD^28,29.

However, other indoles (besides IPA) might influence infection severity and, to the best of our knowledge, no studies have reported the role of indoles in bacterial pneumonia while most severe cases of CAP are caused by bacteria (not influenza)². Furthermore, whether gut microbiota are the main source of systemic indole concentrations is unclear (indoles could also be derived directly from the diet or from host processes) and—most importantly—previous findings await confirmation in patients, which is essential as animal models often overstate the role of microbiota-derived metabolites on extraintestinal immune responses³⁰. Here, using a translational approach of pneumonia mice models combined with general-population and severe CAP-patient cohorts, we investigated the effects of tryptophan metabolites on pneumonia severity. In addition, we assessed the effects of antibiotic-mediated microbiota modulation on systemic tryptophan concentrations in a randomized human intervention trial. We specifically focused on IAA, given its demonstrated protective effects in pancreatic cancer, asthma and COPD^21–23.

Results

Tryptophan metabolites correlate with the risk of hospitalization for pneumonia in the general population

Most CAP patients have mild disease and can be treated on an outpatient basis^2,3. We first wondered whether tryptophan metabolites were related to the probability of having more severe respiratory tract infections (i.e. requiring hospitalization) rather than no or mild disease. We thus used data from 13,464 participants from the large-scale, population-based EPIC-NORFOLK cohort to assess associations between tryptophan metabolites in blood plasma (obtained at study inclusion) and the future occurrence of moderate-severe lower respiratory tract infections, defined as requiring hospitalization or resulting in mortality, during 25−27 years follow-up (Fig. 1A)³¹. In time-to-event analyses (competing risk regression models, treating non-pneumonia mortality as competing risk), the plasma levels of eleven out of fifteen plasma tryptophan metabolites were significantly associated with the risk of hospital admission or mortality due to pneumonia (Fig. 1B). Kynurenine was associated with an increased risk of moderate-severe pneumonia, while IPA was associated with a reduced risk. This aligns with the protective effect of IPA against experimental influenza in mice²⁶. Higher levels of IAA were associated with an increased risk of hospitalization or mortality due to pneumonia, either when IAA was analyzed as a continuous variable (cause-specific hazard ratio [csHR] 1.15, 95% CI 1.09–1.22, Benjamini-Hochberg adjusted p < 0.0001) and when IAA was divided into tertiles to enable visualization (1048 out of 4488 participants with hospitalization or mortality due to pneumonia in the highest tertile versus 958 of 4488 participants in the low tertile: csHR 1.24, 95% CI 1.13–1.35, adjusted p < 0.0001; Fig. 1C). Limiting follow-up to the first 15 years after sample collection, yielded similar results (IAA as continuous parameter: csHR 1.12, 95% CI 1.03–1.23, p = 0.013; low. vs high tertile: csHR 1.28, 95% CI 1.09–1.49, p = 0.0023; Supplementary Table 1). We thus concluded that higher IAA correlated with an increased risk of future hospitalization for pneumonia in the general population.

Fig. 1. Plasma tryptophan metabolite concentrations are associated with the risk of hospitalization or mortality due to pneumonia in the general population.

A Tryptophan metabolites were measured in 13,464 baseline plasma samples of the EPIC-NORFOLK general population cohort. Participants were followed for 25−27 years for the occurrence of moderate-severe lower respiratory tract infections, defined as requiring hospitalization or resulting in mortality. B Forest plot depicting the cause-specific hazard ratios (dots) with 95% confidence intervals (horizontal lines) for the primary outcome (hospital admission or mortality due to a lower respiratory tract infection) of each tryptophan metabolite. Vertical dotted line represents a cause-specific hazard ratio of 1, indicating no association. C Higher baseline levels of plasma indole-3-acetic acid were associated with an increased risk of moderate-severe lower respiratory tract infections, both when indole-3-acetic acid was analyzed as continuous parameter, and when participants were stratified into tertiles based on indole-3-acetic acid levels. Statistical testing was performed by competing risk regression models, treating non-pneumonia mortality as competing risk.

Tryptophan metabolites are altered during severe pneumonia and correlate with mortality

Next, we set out to investigate the association between plasma tryptophan metabolites and the risk of 90 day mortality in patients with severe pneumonia (Fig. 2A). We included 156 patients admitted to the ICU for sepsis due to severe CAP. Patients were part of the large, prospective, observational Molecular Diagnosis and Risk Stratification of Sepsis (MARS) study in two tertiary hospitals in the Netherlands^32–34. A Consolidated Standards of Reporting Trials diagram of study inclusions is depicted in Supplementary Fig. 1. Thirty healthcare workers employed at the Amsterdam University Medical Center (the Netherlands) served as healthy controls^35,36. Table 1 shows the demographic and clinical characteristics of the included patients and controls. Of 156 severe CAP patients, 63 (40.4%) died during the 90 days following ICU admission (Table 1).

Fig. 2. Plasma tryptophan metabolite concentrations are altered during severe pneumonia and correlate with mortality.

A Plasma samples of patients admitted to the ICU with severe community-acquired pneumonia (CAP) were collected within 24 h of ICU admission (n = 156); healthy healthcare workers served as controls (n = 30). Tryptophan metabolites were measured using targeted metabolomics, and associated with 90 day mortality. B Heatmap depicting plasma tryptophan metabolite levels in severe CAP patients, stratified by 90 day mortality (n = 93 survivors and n = 63 non-survivors), and healthy controls. The rightmost columns show effect sizes (Hedges’ g) of comparisons between severe CAP patients and controls, and survivors at day 90 with non-survivors. ‘*’ marks p < 0.05, as tested by two-sided Wilcoxon rank-sum tests. C Schematic representation of measured tryptophan metabolites in severe pneumonia and controls. Each circle represents a metabolite, while the color indicates whether the metabolite is significantly more abundant in pneumonia (red) or controls (blue), or not significantly altered (white). D Schematic representation of measured tryptophan metabolites in severe pneumonia patients, comparing 90 day mortality. Each circle represents a metabolite, while the color indicates whether the metabolite is significantly more abundant in pneumonia patients with mortality within 9 days from ICU admission (red), surviving for at least 90 days (blue), or not significantly altered (white). (E) Higer levels of plasma indole-3-acetic acid at ICU admission were associated with an increased risk of mortality in severe CAP patients (n = 156), both when indole-3-acetic acid was analyzed as continuous parameter, and when stratified into tertiles based on indole-3-acetic acid levels (n = 56 per tertile). Box plots show the median (centerline), the first and third quartiles (the lower and upper bound of the box), and the whiskers show the 1.5 × interquartile range. Statistical testing was performed by two-sided Wilcoxon rank-sum tests and Cox proportional hazards models. Source data are provided as a Source Data file.

Table 1.

Clinical characteristics of the severe CAP cohort

	Severe CAP patients (n =156)	Controls (n = 30)
Age (median, IQR)	63.0 (49.8−72.0)	57.0 (47.3−60.8)
Male sex	98 (62.8)	18 (60.0)
Body Mass Index (median, IQR)1	24.7 (22.0−27.4)	25.4 (23.5−28.0)
Current smoker	25 (16.0)
Comorbidities1
Cardiovascular disease	32 (20.5)	2 (10.5)
Respiratory disease	51 (32.7)	1 (5.3)
Malignancy	22 (14.1)	1 (5.3)
Diabetes	28 (17.9)	0 (0.0)
Renal disease	13 (8.3)	0 (0.0)
Immunocompromised	37 (23.7)	1 (5.3)
Severity of disease at ICU admission
SOFA score	7 (5−10)
APACHE IV score	75 (59−91)
Antibiotics before sample collection
Any	39 (25.0)
Anti-anaerobic antibiotics2	17 (10.9)
Causative pathogen3
No pathogen identified	43 (27.6)
Streptococcus pneumoniae	29 (18.6)
Staphylococcus aureus	21 (13.5)
Haemophilus influenzae	21 (13.5)
Influenza	11 (7.1)
Pseudomonas aeruginosa	11 (7.1)
Escherichia coli	10 (6.4)
Klebsiella species	9 (5.8)
Other Streptococcus species	5 (3.2)
Aspergillus species	2 (1.3)
Other pathogens	33 (21.2)
Outcome
ICU length of stay, days (median, IQR)	5 (2−10)
90 day mortality	63 (40.4)
1 year mortality	79 (50.6)

Data are n (%) unless otherwise indicated. CAP community-acquired pneumonia, IQR interquartile range, ICU intensive care unit, SOFA Sequential Organ Failure Assessment, APACHE Acute Physiology and Chronic Health Evaluation.

¹Data on BMI and comorbidities were collected for n = 19 controls only.

²Piperacillin-tazobactam, meropenem, metronidazole, clindamycin, and amoxicillin with clavulanic acid were considered anti-anaerobic antibiotics.

³In 29 patients ≥2 pathogens were identified; percentages are calculated with the total number of patients as denominator (rather than all pathogens).

We used a validated approach^18,37 to quantify 17 tryptophan metabolites in plasma samples obtained within 24 h of ICU admission. Tryptophan was higher in healthy controls compared to patients with severe CAP (p < 0.0001; Fig. 2B). Eleven downstream metabolites of tryptophan significantly differed between patients and controls, with divergent patterns between and within the three major pathways (indole, serotonin and kynurenine; Fig. 2C). IAA (p < 0.0001), indole-3-sulfate (p = 0.0002) and the sum of all indoles (p = 0.0003) were decreased during CAP, whereas indole-3-aldehyde (p = 0.0020) was higher in patients compared to controls. In addition, the sum of all metabolites from the kynurenine pathway (p = 0.029), and the kynurenine/tryptophan ratio—a common marker the rate-limiting enzyme indoleamine 2,3-dioxygenase-1, which regulates flux of tryptophan into kynurenine¹⁵—were increased in patients (p < 0.0001). Correcting for exposure to any antibiotic prior to sample collection did not affect significance of differences in tryptophan metabolite concentrations between patients and healthy controls, except for 5-OH-indole acetic acid (p = 0.42 after correction). When we corrected for prior exposure to anti-anaerobic antibiotics rather than all antibiotics (considering the detrimental effects of anti-anaerobic antibiotics on gut microbiota composition)^38,39, similar results were obtained

Next, we assessed whether the concentrations of tryptophan metabolites at ICU admission were associated with 90 day mortality. Seven out of seventeen plasma tryptophan metabolites differed between survivors and non-survivors (Fig. 2B, D). Higher total concentrations of the kynurenine and serotonin pathways were associated with an increased risk of mortality at 90 days, aligning with a previously described relationship between high kynurenine and COVID-19 severity, which may reflect a hyperinflammatory response to infection⁴⁰. Indeed, the relationship between kynurenine and mortality was no longer significant when corrected for the archetypal pro-inflammatory cytokines interleukin (IL)−6 or IL-8 (Supplementary Fig. 2), suggesting that kynurenine metabolites primarily reflect hyperinflammation.

Notably, higher levels of IAA at ICU admission were associated with an increased risk of 90 day mortality, both when IAA was analyzed as continuous variable (hazard ratio [HR] 1.31, 95% confidence interval [CI] 1.07-1.62, per log2 increase, p = 0.011), and when we stratified patients into tertiles of low, intermediate, and high IAA (HR 1.83, 95% CI 1.00-3.36, p = 0.051 for low vs. high tertile; Fig. 2E). This aligns with a previously described association between higher levels of IAA and mortality in 60 patients with sepsis⁴¹. Importantly, the association between IAA and mortality in our cohort of severe pneumonia patients was observed even following adjustment for sex, age, body mass index, disease severity (quantified using the Sequential Organ Failure Assessment [SOFA] score), prior antibiotic exposure, causative pathogen and comorbidities (diabetes, malignancy, immunocompromised state, cardiovascular, renal, and respiratory disease) in a multivariable model (HR 1.38, 95% CI 1.06-1.80, p = 0.017; Supplementary Fig. 2). Moreover, when we extended our follow-up and assessed mortality at 1 year following ICU admission, or when we assessed short-term (30 day) mortality, similar effect estimates were found (HR 1.27, 95% CI 1.05-1.53, p = 0.012 for 1 year mortality; HR 1.36, 95% CI 1.07-1.74, p = 0.012 for 30 day mortality), and IAA remained associated with mortality when corrected for IL-6 or IL-8 concentrations (Supplementary Fig. 2). IAA seemed especially related to increased respiratory failure: higher plasma IAA was associated with higher respiratory SOFA scores, and with fewer ventilator-free days (p = 0.033; Supplementary Fig. 3). Conversely, among CAP patients with a bacterial causative pathogen (n = 95), higher IAA was associated with a lower probability of bacteremia with the same pathogen (odds ratio 0.43 per log2 increase of IAA, p = 0.012; Supplementary Fig. 3).

Taken together, most indoles decrease during severe CAP, while tryptophan metabolites from the kynurenine pathway are higher. In patients with severe CAP, IAA is associated with higher mortality, independent from potential confounders.

Effects of broad-spectrum antibiotics on tryptophan metabolites in a human intervention trial

Previous studies have reported that indole concentrations are decreased in germ-free mice and are affected by microbiota transfer experiments^17,21, aligning with the presumed gut microbiota-derived nature of indoles. Yet, limited human data is available while the gut microbiota composition of humans vastly differs from mice, and metabolites of tryptophan could also be derived from the diet or by bacteria colonizing the respiratory tract. Therefore, we next investigated the effects of gut microbiota modulation by broad-spectrum antibiotics on plasma tryptophan metabolite levels through a human intervention trial. Details on the study design, participant characteristics and microbiota characterization have been previously published⁴². In short, 20 participants (aged 18−45 years) with intermittent-to-mild asthma were recruited and randomized into two groups, which received either no treatment or oral broad-spectrum antibiotics (ciprofloxacin 500 mg every 12 h, vancomycin 500 mg every 8 h and metronidazole 500 mg every 8 h) for 7 days in order to disrupt the gut microbiota. Following a 36 h washout period, blood plasma was collected and a bronchoalveolar lavage was performed (Fig. 3A). The 7 day course of broad-spectrum antibiotics resulted in a drastic shift in gut microbiota community composition (β-diversity, p = 0.0001; Fig. 3B), as previously described⁴². Participants who received broad-spectrum antibiotics had decreased abundances of several obligate anaerobic bacteria (e.g. Ruminococcaceae, Bifidobacterium and Blautia) and overgrowth of Streptococcus, Lactococcus and Lactobacillus (Fig. 3C). Antibiotic-mediated microbiome modulation resulted in decreased plasma concentrations of IAA, indole-3-sulfate, total indoles and xanthurenic acid whereas 3-OH-kynurenine was increased (Fig. 3D). Specifically, broad-spectrum antibiotics resulted in a 40% decrease in median plasma IAA concentrations (609.6 versus 365.3 pmol/mL). Of the genera altered by antibiotic treatment, none were significantly associated with concentrations of tryptophan metabolites (Benjamini-Hochberg adjusted p > 0.05). In the lungs (i.e. bronchoalveolar lavage fluid), most tryptophan metabolites were below the lower limit of detection, both in participants who received antibiotics and controls (Supplementary Table 2)—indicating that the lungs are not a relevant source of plasma concentrations of tryptophan metabolites. Thus, this human interventional trial shows that broad-spectrum antibiotic treatment results in decreased plasma concentrations of indoles, suggesting that gut microbiota are an important modulator of systemic indoles in humans.

Fig. 3. Effects of microbiota modulation by broad-spectrum antibiotics on plasma tryptophan metabolites in a human intervention trial.

A Twenty participants were randomized to receive either no treatment (control; n = 13) or oral broad-spectrum antibiotics for 7 days in order to disrupt the gut microbiota (n = 7). Following a 36 h washout period, blood plasma was collected in which tryptophan metabolites were measured using targeted metabolomics. B Broad-spectrum antibiotics resulted in altered gut microbiota community composition. Significance of differences in community composition between groups is determined using permutational multivariate ANOVA (two-sided) with Bray-Curtis dissimilarities at the level of amplicon sequence variants. C This compositional difference between groups was driven by higher relative abundances of Streptococcus, Lactococcus, and Lactobacillus spp. in participants who received broad-spectrum antibiotics, and lower relative abundances of several obligate anaerobic bacteria (e.g. Ruminococcaceae UCG-002 and Bifidobacterium spp.), as identified by a DESeq2 model. D Heatmap depicting plasma tryptophan metabolite levels in the participants with broad-spectrum antibiotic-mediated gut microbiota modulation and controls. The rightmost columns show effect sizes (Hedges’ g) of comparisons between the antibiotics group and controls. ‘*’ marks p < 0.05, as tested by two-sided Wilcoxon rank-sum tests. Source data are provided as a Source Data file.

Dynamics of tryptophan metabolites during experimental bacterial pneumonia

Since tryptophan metabolites were associated with the risk of pneumonia in the general population, altered during hospitalization for severe CAP, and previous studies showed that tryptophan metabolism is affected by inflammation¹⁵, we aimed to establish whether pneumonia itself induces changes in tryptophan metabolite. To fully control for variation in environmental and patient-related factors (e.g. demographics and treatments), we used our well-established mouse model of severe bacterial pneumonia^43–45 to assess the dynamics of tryptophan metabolites: C57Bl/6 J mice were intranasally inoculated with Klebsiella (K.) pneumoniae to evoke pneumonia, and plasma was harvested directly following infection, and after 12, 24 and 36 h (Fig. 4A). K. pneumoniae is the most common Gram-negative cause of pneumonia and associated with higher mortality compared to most other bacterial pathogens^46,47. Dynamics of bacterial counts, inflammatory cytokines and microbiota have been previously published⁴⁰. During experimental pneumonia, eleven out of seventeen tryptophan metabolites were significantly altered in plasma (Fig. 4B). Tryptophan increased, while stark differences existed between the three downstream tryptophan pathways. Plasma concentrations of several indoles (e.g. IAA, indole-3-sulfate) and metabolites of the serotonin pathway (e.g. serotonin, 5-OH-indole acetic acid) decreased during pneumonia, whereas the kynurenine-related metabolites (e.g. kynurenine, quinolinic acid) increased (Fig. 4B, C). Tryptophan metabolites were lower in lungs compared to plasma, yet had largely similar dynamics (Supplementary Fig. 4). Specifically, pulmonary IAA concentrations were below the lower limit of detection (<50 pmol/ml) before and during K. pneumoniae-induced pneumonia. This aligns with the low pulmonary concentrations in participants from our human interventional trial and suggests that lung microbiota (including K. pneumoniae) were not a relevant source of plasma IAA. We concluded that in both mice and patients with pneumonia (compared to healthy controls), total kynurenine metabolites increased and indoles decreased, indicating that bacterial pneumonia induces repartitioning of tryptophan metabolism.

Fig. 4. Dynamics of plasma tryptophan metabolites during experimental bacterial pneumonia.

A C57BL/6 J mice were intranasally infected with Klebsiella pneumoniae to evoke bacterial pneumonia, and plasma was harvested directly following infection, and after 12, 24 and 36 h (n = 10 mice per timepoint). Tryptophan metabolites were quantified in blood plasma. B Forest plot depicting the change of each tryptophan metabolite over time during pneumonia (the dot represents the estimate and the line ±2 standard deviations). Vertical dotted line represents an effect size of 0, indicating no change. Statistical testing is performed using linear mixed models. C Graphical display of the dynamics from eight selected tryptophan metabolites during experimental bacterial pneumonia. Dots depict the mean, with vertical bars showing the standard error of the mean. Source data are provided as a Source Data file.

IAA aggravates pulmonary damage and reduces dissemination during bacterial pneumonia

We next investigated the effects of IAA during experimental pneumonia. We focused on IAA since it was the only indole consistently associated with pneumonia severity in both the general population and ICU patients. In addition, IAA was (in contrast to all other metabolites except for 5-OH-indole acetic acid) associated with higher mortality during severe pneumonia, independent from clinical confounders and pro-inflammatory cytokines, and was among the few tryptophan metabolites that were significantly altered following gut microbiota modulation. Moreover, IAA supplementation has protective effects against inflammatory bowel disease and COPD^18,22, and enhances the efficacy of chemotherapy in pancreatic cancer²¹. Consequently, IAA supplementation was suggested as novel treatment for these diseases^28,29. The contrast between these protective effects of IAA, and the association of IAA with more severe pneumonia in our cohorts, served as additional impetus to further investigate IAA. We specifically wondered whether IAA pretreatment would increase pulmonary damage and reduce bacteremia, which would mirror the association between higher IAA levels and respiratory SOFA scores and fewer positive blood cultures in patients with severe CAP. We pretreated C57Bl/6 J mice for 2 weeks with IAA by daily intraperitoneal injections. Dosage and route were based on previously published studies demonstrating beneficial effects of IAA in models of COPD and nonalcoholic fatty liver disease^22,48. IAA pretreatment increased plasma IAA concentrations by ~50% (median 316.3 versus 210.1 pmol/mL, p = 0.015; Supplementary Fig. 5). Following pretreatment, pneumonia was induced through intranasal administration of K. pneumoniae, as described^43–45. Mice were sacrificed 24 or 42 h after infection to assess outcomes (Fig. 5A). Two of the IAA-pretreated mice died prematurely (~36 h following infection; 25% of those not yet sacrificed at 24 h), compared to none of the control mice, suggestive of harm induced by IAA. IAA pretreatment resulted in an increase in pulmonary K. pneumoniae burden at 24 h (p = 0.035), though not significant at 42 h following infection (p = 0.14; Fig. 5B). Examination of lung histology by a blinded pathologist revealed that IAA-pretreated mice had significantly more pulmonary damage (interstitial inflammation, endothelialitis, bronchitis and edema) at 24 h following K. pneumoniae-induced pneumonia (p = 0.043; Fig. 5C, D). This difference was even more pronounced at 42 h (p = 0.014). Of note, IAA pretreatment did not result in pulmonary damage when mice were not subjected to experimental pneumonia (Supplementary Fig. 5). Conversely, mice pretreated with IAA had lower bacterial counts in their blood (Fig. 5E) and reduced distant organ damage as illustrated lactate dehydrogenase (LDH), and the hepatic injury markers aspartate aminotransferase (AST) and alanine aminotransferase (ALT; Fig. 5F). A single day of IAA pretreatment prior to infection did not affect outcomes of bacterial pneumonia (Supplementary Fig. 6). We thus concluded that 14 day IAA supplementation had two main effects during bacterial pneumonia: IAA aggravated lung damage, while reducing systemic dissemination of bacteria, resembling findings in severe CAP patients.

Fig. 5. IAA aggravates pulmonary damage and reduces dissemination during bacterial pneumonia.

A C57BL/6 J mice were pretreated for 2 weeks with saline (control) or indole-3-acetic acid (IAA), and subsequently intranasally infected with Klebsiella pneumoniae to evoke pneumonia. After 24 and 42 h, bacterial outgrowth, organ injury and organ damage parameters were determined (n = 8 mice per group and timepoint; two IAA-pretreated mice died before the 42 h endpoint and were not analyzed). B Effects of IAA-pretreatment on bacterial counts in lungs following infection with K. pneumoniae. C Pulmonary damage was higher in IAA-pretreated mice compared to controls at both timepoints. D Representative lung slides of control (upper) and IAA-pretreated (lower) mice infected with K. pneumoniae and euthanized at 42 h (H&E staining; original magnification × 10). E Bacterial counts in the blood after K. pneumoniae infection. F Plasma lactate dehydrogenase (LDH), aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels. Bars represent the mean with standard deviation. Statistical testing was performed by two-sided Wilcoxon rank-sum tests. Source data are provided as a Source Data file.

IAA impacts bacterial pneumonia through AhR

Considering the important role of AhR signaling in immune responses and that IAA serves as a ligand of AhR^15,16,22,27, we next sought to determine whether effects of IAA on K. pneumoniae-induced pneumonia were AhR-dependent. Mice were pretreated with saline (control), IAA, or both IAA and an AhR-inhibitor (CH-223191) by intraperitoneal injections²². Following 2 weeks of pretreatment, mice were intranasally inoculated with K. pneumoniae and sacrificed at 24 h following infection (Fig. 6A). No effects of either pretreatment on pulmonary bacterial outgrowth were observed (Fig. 6B). However, the IAA-induced aggravation of lung damage (compared to controls: p = 0.025) was largely abrogated in mice simultaneously treated with IAA and the AhR-inhibitor (p = 0.91 compared to controls; Fig. 6C), suggesting that the effects of IAA on pulmonary damage were at least partly AhR-dependent. Similarly, IAA reduced bacterial loads in the blood, liver and spleen of mice with bacterial pneumonia compared to controls, whereas these effects of IAA fully disappeared in the presence of the AhR-inhibitor (Fig. 6D), showing that IAA reduced bacterial dissemination in an AhR-dependent manner. Yet, when mice were pretreated with the AhR agonist FICZ (6-Formylindolo[3,2-b]carbazole) or with indole-3-lactic acid (ILA), which also serves as AhR ligand, we did not observe any effects of either pretreatment on bacterial pneumonia (Supplementary Fig. 7), underlining a specific role of IAA.

Fig. 6. IAA impacts K. pneumoniae-induced pneumonia through AhR.

A C57BL/6 J mice were pretreated for 2 weeks with saline (control), indole-3-acetic acid (IAA), or both IAA and an inhibitor of the aryl hydrocarbon receptor (AhR; CH-223191, n = 8 mice per group). Following pretreatment, mice were subjected to intranasal infection with Klebsiella pneumoniae, and euthanized after 24 h to assess outcomes. B Bacterial counts in the lungs following bacterial pneumonia did not differ between pretreatment groups. C Pretreatment with IAA resulted in more pulmonary damage compared to controls, which was largely abrogated in mice pretreated with both IAA and the AhR-inhibitor. D Bacterial counts in the blood, liver and spleen following infection with K. pneumoniae were lower in IAA-pretreated mice, but not when simultaneously pretreated with an AhR-inhibitor. Bars represent the mean with standard deviation. Statistical testing was performed by two-sided Wilcoxon rank-sum tests. Source data are provided as a Source Data file.

IAA increases ROS release during in vitro K. pneumoniae infection and in patients

We next aimed to identify the mechanism underlying the detrimental effects of IAA on lung damage during bacterial pneumonia. Neutrophil influx, pulmonary cytokine levels (TNF, IL-1β, IL-6, IL-10, monocyte chemoattractant protein-1, chemokine CXC motif ligand 1) and plasma cytokine concentrations (TNF, IL-6) did not differ between IAA-pretreated mice and controls during bacterial pneumonia (Supplementary Fig. 8). Similarly, in severe CAP patients, plasma IAA concentrations were not associated with the plasma concentrations of cytokines or inflammatory markers (p > 0.05 for all 20 measured host immune response biomarkers; Supplementary Fig. 9).

Previous studies described that IAA increases ROS release from cultured neutrophils^21,49. Extracellular ROS are important in the defense against bacterial outgrowth during pneumonia, yet simultaneously induce pulmonary damage through oxidative stress^50,51. We thus wondered whether the effect of IAA on aggravated lung damage was mediated by ROS. We pretreated mice with IAA for 2 weeks, and subsequently employed a previously described assay with isoluminol to detect extracellular O_2- production by neutrophils⁵⁰. Neutrophils from IAA-pretreated mice produced more extracellular ROS following stimulation with K. pneumoniae compared to controls (Fig. 7A, B). In line, IAA pretreatment increased oxidative stress during K. pneumoniae-induced pneumonia in mice, as quantified by pulmonary nitrotyrosine levels (Supplementary Fig. 10). This suggests that IAA aggravates pulmonary damage by increasing ROS release, resulting in more oxidative stress. The IAA-induced increase in oxidative stress was considerably abrogated in mice simultaneously treated with IAA and an AhR-inhibitor (Supplementary Fig. 10).

Fig. 7. IAA increases ROS production during in vitro K. pneumoniae infection and in patients.

Reactive oxygen species (ROS) release from bone-marrow derived neutrophils of mice pretreated for 2 weeks with saline (control) or indole-3-acetic acid (IAA; n = 6 mice per group). ROS was quantified by isoluminol chemiluminescence after in vitro stimulation with live Klebsiella pneumoniae (multiplicity of infection of 10). Dots represent the mean and bars the standard error of the mean (only upper half is shown to avoid overlap). B Neutrophils from IAA-pretreated mice (n = 6) produced more extracellular ROS (calculated as the area under the curve from panel (A) compared to controls (n = 6), both following stimulation with K. pneumonia and phorbol myristate acetate (PMA). Bars represent the mean with standard deviation. Statistical testing was performed by two-sided Wilcoxon rank-sum tests. C In a subset of severe CAP patients (n = 66), transcriptomics was performed on whole blood samples collected within 24 h after ICU admission. Higher plasma IAA levels were associated with upregulation of pathways depicted in red. Green bars represent pathways significantly associated with lower IAA levels. Reactome pathways related to ‘Immune System’ were assessed. D Higher IAA levels correlated with increased expression of genes involved in the Reactome pathway ‘ROS and RNS production in phagocytes’. Source data are provided as a Source Data file.

Finally, we wanted to assess the relevance of the association between IAA and increased ROS in patients with severe pneumonia. We used whole blood transcriptomics data, collected within 24 h of ICU admission, from a subset of our severe CAP cohort (n = 66; those enrolled during the first 1.5 years)³⁴. Higher plasma IAA (analyzed as continuous variable) correlated with increased expression of genes involved in the Reactome pathway ‘ROS and RNS production in phagocytes’ (normalized enrichment score 1.76, Benjamini-Hochberg adjusted p = 0.023; Fig. 7C, D). In addition, IAA was associated with upregulation of pathways involved in neutrophil degranulation and cellular responses to stress, corresponding with the increase in tissue damage (Fig. 7C). Lower IAA correlated with upregulation of genes involved in interferon signaling, aligning with earlier observed lower interferon levels in indole-treated mice²⁶. Together, we concluded that IAA aggravates pulmonary damage during pneumonia in mice and humans by a mechanism that likely involves increased ROS release from neutrophils.

Discussion

In this translational study—combining population-based and severe CAP cohorts with murine pneumonia experiments—we found that the gut microbiota-derived tryptophan metabolite IAA affects pneumonia and modulates interpatient heterogeneity in pneumonia. IAA was associated with an increased risk of future hospital admission for pneumonia among 13,464 general population subjects, and with higher mortality during severe CAP. In a randomized human intervention trial, microbiota modulation by broad-spectrum antibiotics resulted in decreased plasma IAA concentrations. Further, we found that IAA supplementation aggravated pulmonary damage while reducing systemic dissemination during K. pneumoniae-induced pneumonia in mice, via AhR and by increasing ROS. Based on these observations, we conclude that while IAA may increase immune resistance (i.e., limiting pathogen dissemination), it comes with the consequence of reduced tissue resilience (processes striving to limit tissue damage)^2,52. This is, to the best of our knowledge, the first assessment of tryptophan metabolites in severe pneumonia and the susceptibility to lower respiratory tract infections.

Gut microbiota are an important modulator of immune responses during pneumonia through the production of immunomodulatory metabolites^5,13,14. While much is known about the role of short-chain fatty acids, other microbiota-derived metabolites are underappreciated. If other metabolites are studied, attention focuses on intestinal rather than extraintestinal pathophysiology. Here, we aimed to address these knowledge gaps. In a general population cohort, baseline kynurenine, kynurenic and quinolonic acid were higher among those who developed moderate-severe lower respiratory tract infections during follow-up. Furthermore, multiple indoles, including IAA, were associated with pneumonia susceptibility. Similarly, in patients with severe CAP, higher plasma IAA was predictive of increased mortality, which was robust after adjusting for age, sex, body mass index, disease severity, causative pathogen, and comorbidities. Similarly, several kynurenine metabolites were associated with mortality during severe CAP although this probably primarily reflects hyperinflammation. Since IAA was one of two tryptophan metabolites associated with higher mortality during severe pneumonia independent from clinical confounders and hyperinflammation (the other being 5-OH-indole acetic acid), and the only indole consistently associated with pneumonia severity in both the general population and ICU patients, we aimed to specifically elucidate the causal effects of IAA during pneumonia. We used our well-established model of K. pneumoniae-evoked severe pneumonia, which mimics human pneumonia-derived sepsis^43–45. Pretreatment of mice with IAA increased pulmonary damage, while reducing bacterial dissemination. This aggravated pulmonary damage corresponds with the positive association between IAA and pulmonary injury in severe CAP patients. Furthermore, the IAA-induced reduction in systemic bacterial counts matches the observed association between increased IAA and a lower risk of bacteremia, translating the murine evidence to patients. Together, our findings suggest that IAA results in increased lung injury during pneumonia, potentially leading to a higher likelihood of requiring hospitalization in the general population, and an increased risk of mortality during severe CAP. The detrimental effect of IAA-pretreatment on pneumonia is in contrast with previous studies describing protective effects against inflammatory bowel disease, pancreatic cancer and pulmonary diseases (COPD and asthma)^18,21–23. Specifically, the increased pulmonary damage diverges from the attenuation of tissue injury by IAA (in the same dose) in a murine COPD model²²,suggesting that IAA has differing roles in pulmonary diseases. Moreover, effects of other indoles might be different¹⁶: IPA protected against murine influenza-induced pneumonia and abdominal-derived sepsis^25,26. Although we did not measure IPA in severe CAP, IPA was associated with a reduced risk of lower respiratory tract infection hospitalization in the general population, whereas other indoles (e.g., indole-3-lactic acid, IAA) were associated with a higher risk. In line, both ILA and the AhR agonist FICZ did not affect experimental pneumonia in mice. This implies that different indoles may have opposite effects during systemic infections, and further adds to the complexity of microbial metabolite-host interactions.

Pneumonia itself also affected tryptophan metabolism. In patients with severe CAP, tryptophan metabolite concentrations were extensively altered at ICU admission: plasma tryptophan and indoles were lower, while the kynurenine pathway was increased compared to healthy controls. Experimental pneumonia in mice induced similar changes: a decrease of indoles and increased kynurenine metabolites, suggesting that alterations in patients are (at least partly) induced by the infection, rather than preceding pneumonia onset. An increase in kynurenine metabolites is observed in many inflammatory diseases. The flux of tryptophan into kynurenine is mediated by the rate-limiting enzyme indoleamine 2,3-dioxygenase-1, which is induced by pro-inflammatory cytokines (e.g. TNF)^15,53. Upregulation of indoleamine 2,3-dioxygenase-1 and enhanced kynurenine production during inflammation may result in decreased availability of tryptophan, resulting in lower levels of indoles¹⁵. Our findings confirm this repartitioning of tryptophan fluxes during pneumonia. In line, tryptophan metabolites were commonly altered at 24 h post-infection while unaffected at 12 h, corresponding with the infection (and inflammation) expanding from the lungs to the systemic circulation at 12–24 h after inoculation⁴³. In addition, pneumonia alters gut microbiota composition, potentially through infection-induced lack of appetite⁷, which could further contribute to the differences in tryptophan metabolites during pneumonia. Yet, it is unclear whether changes in food-seeking behavior and gut microbiota of mice contribute to the alterations in tryptophan metabolism observed here.

We found that gut microbiota significantly contributed to systemic indole concentrations in humans. In a randomized human intervention trial, we demonstrated that indoles are reduced following modulation of microbiota through broad-spectrum antibiotics, strongly suggesting a microbial origin of systemic indoles in humans although we could not definitively exclude dietary or host enzymatic sources of indoles. In the lungs, IAA concentrations were below the lower limit of detection which implies that pulmonary microbiota are not a relevant source of plasma IAA. We could not identify individual intestinal genera significantly associated with tryptophan metabolites. Previous studies reported that IAA is produced by diverse bacteria, including facultative anaerobic Lactobacillus species and Escherichia coli, and obligate anaerobic Bacteroides spp^21,54,55. Interestingly, increased abundances of E. coli have been associated with adverse clinical outcomes during respiratory infections and critical illness whereas obligate anaerobic are associated with lower mortality^38,39,56,57. Future studies may elucidate whether these associations are mediated by IAA. Of note, it has recently been shown that substrate (e.g., dietary fiber) availability rather than the abundance of tryptophan-metabolizing bacteria directs the metabolism of tryptophan into different indoles⁵⁸.

Mechanistically, the effects of IAA on bacterial pneumonia were—at least partly—AhR-dependent. Indoles are essential in maintaining intestinal barrier function and, more relevantly, AhR signaling enhanced the lung barrier during influenza-induced pneumonia^15,27,59. Thus, AhR activation may have strengthened pulmonary barrier function and prevented spreading of Klebsiella from the lungs into the blood. Future studies may further elucidate the effects of IAA on bacterial dissemination. Notably, the AhR agonist FICZ did not affect lung damage or bacterial dissemination, nor did the AhR ligand ILA, suggesting a convoluted role of IAA and AhR signaling in pneumonia. AhR activation commonly dampens inflammation^15,53, yet IAA-pretreatment did not impact cytokine responses or airway neutrophilia during experimental pneumonia, nor was IAA associated with host immune response biomarkers in severe CAP patients. Instead, IAA increased ROS release from murine neutrophils, induced pulmonary oxidative stress, and was associated with increased expression of genes involved in ROS production by blood leukocytes in severe CAP patients. ROS is a major determinant of tissue damage during pneumonia⁵¹, and our findings add to previous studies showing that IAA increases ROS from neutrophils, resulting in tissue damage^21,49. Our study is the first to assess the effects of IAA-pretreatment during bacterial pneumonia, and highlights the role of IAA-induced ROS release in tissue resilience.

This study has several shortcomings. First, owing to the observational design of the human cohorts, the relations between tryptophan metabolites and pneumonia severity may be affected by confounding. Yet, the association between IAA and mortality remained significant in multivariable models that controlled for confounders, and key effects of IAA overlapped between patients and mice (e.g., increased pulmonary injury and reduced risk of bacteremia). Furthermore, tryptophan metabolites were measured at a single timepoint in our human cohorts while levels might change over time. Second, we employed a commonly used murine model of severe bacterial pneumonia by intranasal infection with K. pneumoniae. This well-established model mimics key pathophysiological aspects of pneumonia-derived sepsis in humans^43–45, but may not entirely represent the full heterogeneity of CAP in humans. In addition, we and others have shown that gut microbiota of genetically similar mice differ between vendors and shipments^60,61, which may have influenced findings obtained in murine experiments. Finally, future studies may identify additional mechanisms, such as effects on airway epithelium or regulatory T cells, involved in the relationship between IAA and pneumonia severity.

In summary, we identified IAA as an important contributor to heterogeneity in CAP severity. In translational analyses, IAA increased pulmonary damage while reducing systemic dissemination via AhR-activation and increasing the release of reactive oxygen species from neutrophils.

Methods

Ethics statement

Ethical approval for the EPIC-Norfolk study was obtained from the Norfolk Research Ethics Committee³¹. The S3 study was approved by the local Medical Ethical Review Board and accepted by the competent authority, the Central Committee on Research on Human Subjects (NL73478.029.20). For both studies, written informed consent was obtained from all participants. For the MARS cohort, inclusion of patients was done using an opt-out method approved by the institutional review boards of the Academic Medical Center Amsterdam and the University Medical Center Utrecht (IRB No. 10-056 C). The Medical Ethics Committee of the Academic Medical Center, Amsterdam (the Netherlands), approved the CAST study (reference: NL52003.018.15) and all patients gave written informed consent. For all studies, this included permission for the analyses performed and consent to publish our findings. All research was conducted in accordance with the Declaration of Helsinki. Animal experiments were conducted in accordance with the Dutch Experiment on Animals Act and European Directives, and approved by the Central Authority for Animal Experiments and the Animal Welfare Body of the Amsterdam UMC.

EPIC-NORFOLK cohort

To assess the relationship between plasma tryptophan metabolites and incident severe (i.e. either hospitalization or mortality) lower respiratory tract infections in the general population, we used data from the population-based European Prospective Investigation into Cancer (EPIC)-Norfolk cohort. Details on the study design, in- and exclusion criteria, measurement of metabolites, and usage of national registries for clinical outcomes have been previously described^31,62. In brief, the EPIC-Norfolk cohort included middle-aged participants from the general population of Norfolk (United Kingdom), of whom 13,464 were selected for metabolomic profiling. Non-fasted plasma samples were collected at baseline in 1993−1997, and stored in liquid nitrogen. The Discovery HD4 platform (Metabolon) was used for untargeted metabolomic measurements, details of which have been previously described³¹. Metabolite levels were normalized and rescaled to a mean of zero and standard deviation of one³¹. For the purposes of this study, fifteen tryptophan metabolites were investigated; other metabolites were not analyzed. Our primary outcome was hospital admission or mortality due to a lower respiratory tract infection during follow-up after collection of a plasma sample at study inclusion. We opted for this outcome to identify the more severe cases of pneumonia, rather than those that could be treated on an outpatient basis. Participants were linked to national hospitalization and mortality registries as previously described³¹. Patients were identified as having reached the primary outcome if one of the following ICD-9 or ICD-10 codes was registered as reason for hospitalization or cause of death: 466, 480-487, 513, J10-J22, and J85. Mortality from other causes was treated as competing risk.

Severe CAP patient cohort

Patients with severe community-acquired pneumonia (CAP) were included as part of the Molecular Diagnosis and Risk Stratification of Sepsis (MARS) study (ClinicalTrials.gov identifier NCT01905033). Details on the study design, inclusion criteria and procedures have been previously described^32–34. In brief, MARS was a prospective observational in the ICUs of the Academic Medical Center (Amsterdam) and the University Medical Center Utrecht, both in the Netherlands, conducted between January 2011 and January 2014. Based on prospectively collected data during admission, the likelihood of CAP (and other infections) at ICU admission was post hoc labeled as ‘none’, ‘possible’, ‘probable’, or ‘definite’ by an adjudication committee, as previously detailed⁶³. In addition, patients were post hoc labeled as having sepsis (based on fulfilling the Sepsis-3 criteria⁶⁴). For the current study, patients admitted to the ICU with ‘probable’ or ‘definite’ CAP and sepsis were included. Patients transferred from other ICUs, with other concurrent infections, or no sepsis within 24 h of ICU admission were excluded. For patients with multiple admissions for sepsis, only the first was included. To enable the measurement of tryptophan metabolites and assess correlations with host immune response biomarkers, we included those patients of whom citrate plasma was collected within 24 h of ICU admission, and host response biomarkers were measured (i.e. enrolled within the first 2.5 years of the MARS project). Sex was determined based on self-report or as captured in the electronic health records. The causative pathogen of included patients was determined based on all available microbiology results. Our primary clinical outcome for severe CAP patients was mortality at 90 days after ICU admission. Sensitivity analyses included outcome assessment at an earlier (30 days) and later timepoint (1 year). To assess the relationship between IAA and respiratory failure, we used the highest respiratory component of the SOFA score, and calculated the number of ventilator-free days (i.e., alive and not mechanically ventilated), both adjudicated 90 days after ICU admission. Among severe CAP patients with a bacterial causative pathogen, bacteremia was considered present when the pathogen causing the pneumonia was (also) detected in a blood culture.

Healthcare workers employed at the Amsterdam University Medical Center and included in the prospective observational S3 study served as healthy controls (Netherlands Trial Register NL8645). Details have been previously described^35,36. Data on baseline BMI and comorbidities were collected after the initial study enrollment and thus not available for all participants. Citrate plasma, obtained prior to SARS-CoV-2 vaccination, of healthcare workers of similar age and sex (self-reported) as the patient cohort, was used for quantification of tryptophan metabolites.

Quantification of tryptophan metabolites

Blood citrate plasma, BALF and murine lung homogenate samples were shipped to the University of Tours (UMR 1253, iBrain, University of Tours, Inserm, France) to measure tryptophan metabolites using a previously described and validated method^18,37. In brief, liquid chromatography coupled with high resolution mass spectrometry was used to quantify absolute amounts of tryptophan and sixteen tryptophan metabolites. For each compound, a calibration curve was created by calculating the intensity ratio obtained between the metabolite and its internal standard. The total amount of indoles was calculated by summing the concentrations of IAA, indole-3-aldehyde, indole-3-lactic acid, and indole-3-sulfate. Similarly, the sums of metabolites from the serotonin and kynurenine pathway were calculated. For both severe CAP patients and healthy controls, the quantification and analysis of tryptophan metabolites were not part of a preregistered analysis plan.

Whole blood transcriptomics

Details on whole blood transcriptomics have been previously described^32–34,65. In brief, in a subset of severe CAP patients (n = 66; those enrolled during the first 1.5 years of the MARS study), whole blood was collected in PAXgene tubes (Becton-Dickinson, Breda, the Netherlands) within 24 h of ICU admission, and stored at −80 °C. The PAXgene blood mRNA kit (Qiagen, Venlo, the Netherlands) was used to isolate total RNA. RNA (with an integrity number ≥6) was processed and hybridized to the Affymetrix Human Genome U219 96-array, and scanned using the GeneTitan instruments at the Cologne Center for Genomics (Cologne, Germany).

Host immune response biomarkers measurements

We employed a Luminex multiplex assay (R&D Systems Inc, Minneapolis, MN) on a BioPlex 200 (BioRad, Hercules, CA) to measure 20 host immune response biomarkers reflective of inflammatory cytokines, inflammation and organ damage, and the endothelial and coagulation response in severe CAP patients, as earlier detailed^34,65–67. All host immune response biomarkers were measured in EDTA anti-coagulated plasma obtained within 16 h after ICU admission.

Human intervention trial

This proof-of-concept, single center, randomized, open-label, controlled intervention trial was part of the CAST (C1-inhibitor in allergic ASThma patients) study, in which the effect of C1-inhibitor or microbiota depletion on lung inflammation in asthma patients with allergy was investigated. The study protocol is registered at ClinicalTrials.gov (identifier: NCT03051698) on 14 February 2017. Details on the study design, screening procedures, in- and exclusion criteria and participant characteristics have been previously published^42,68. In short, patients (aged 18−45 years) with intermittent-to-mild asthma according to criteria of the Global Initiative for Asthma, sensitization to house dust mite and no clinically significant abnormalities during physical examination, hematological and biochemical screening were included. Following inclusion, subjects were randomly assigned to either broad-spectrum antibiotics or no antibiotic control. All participants collected a fecal sample at baseline, which was stored at −20 °C at home and transported to the study center for storage at −80 °C within 24 h. Participants allocated to the antibiotics group received 7 days of oral broad-spectrum antibiotics (ciprofloxacin, 500 mg every 12 h; vancomycin, 500 mg every 8 h; and metronidazole, 500 mg every 8 h). This antibiotic regimen is similar to regimens previously used by our group^69,70, and chosen based on the assumption that extensive gut microbiota alterations would make the effects of gut microbiota most evident. After a 36 h washout period (without antibiotic administration—to avoid direct interference of the antibiotics with any of the subsequent measurements), blood was drawn from all participants, and participants from the antibiotics group collected a second fecal sample (storage was performed as described above). In addition, as detailed earlier^42,68, a bronchoscopy was performed in all participants to instill one lung segment with house dust mite and lipopolysaccharide, and saline in the contralateral segment. During a second bronchoscopy 7 h later, both instilled lung segments were lavaged with 8 successive 20 mL aliquots of saline. Bronchoalveolar lavage fluid (BALF) from both instilled lung segments was centrifuged (400 g, 10 min, 4 °C) and supernatant was stored at −80 °C until further analysis. We quantified tryptophan metabolites in BALF from the saline-challenged lung segment. Although the modulation and analysis of gut microbiota were pre-specified in the study protocol, quantification of tryptophan levels in plasma and BALF were not part of the pre-specified analyses. All outcomes of pre-specified analyses have been previously published^42,68.

Gut microbiota analysis

Bacterial microbiota were analyzed as previously described by our group^10,42. In summary, DNA was extracted by a repeated bead beating protocol. Next, DNA was purified with the Maxwell RSC Blood DNA Kit (Promega, Madison, WI) and eluted in 50 μL DNAse free water. Twenty nanograms of DNA were used for the amplification of the 16S rRNA gene with the V3-V4 341 F forward primer (5’-CCTACGGGNGGCWGCAG-3’) and the 805 R reverse primer (5’-GACTACHVGGGTATCTAATCC-3’) for 25 cycles. Amplified product was purified using AMPure XP beads (Beckman Coulter, Indianapolis, IN; according to manufacturer’s guidelines). Purified products were equimolar mixed and 250 bp paired-end sequenced with 2 × 251 cycles on an Illumina MiSeq platform (GATCBiotech, Constance, Germany) using V3 chemistry, according to manufacturer’s instructions. Sequence reads were analyzed as follows. Read pairs with perfect matching forward and reverse barcodes were assigned to their corresponding samples. The forwards and reverse reads were length trimmed at 240 and 210, respectively, which were inferred and merged with ASVs using DADA2 (V1.5.2). The assignment of taxonomy was done using the DADA2 implementation of the RDP classifier and SILVA 16S reference database. Relative abundances were used throughout this study. Differences in overall microbiota composition (β-diversity) were assessed by permutational ANOVA using Bray-Curtis dissimilarities (adonis function, vegan package, 9999 permutations), and visualized using principal coordinates analysis. DESeq2 was used to test for differentially abundant taxa (restricted to genera present with at least 100 reads in at least 20% of the participants).

Mice experiments

Specific pathogen-free female C57BL/6 J mice were purchased from Charles River, and housed in individually ventilated cages in rooms with a controlled temperature (20–26 °C with 30–70% humidity) and 12 h light-dark cycle at the Animal Research Institute Amsterdam (ARIA) facility of the Academic Medical Center facility under standard care. Mice were acclimatized for 1 week prior to experiments. All experiments were conducted with mice between 9 and 11 weeks of age at time of infection.

IAA, AhR-inhibitor, FICZ and ILA pretreatment

Mice were pretreated by daily intraperitoneal injections with 50 mg/kg bodyweight indole-3-acetic acid (Sigma Aldrich) for 2 weeks. Control mice received daily intraperitoneal injections with saline. For experiments with AhR inhibition, CH-223191 (Sigma Aldrich) was given by daily intraperitoneal injections (10 mg/kg bodyweight) for 2 weeks, prior to IAA administration. ILA (50 mg/kg bodyweight; Sigma Aldrich) and FICZ (100 µg/kg bodyweight; Sigma Aldrich) were also administered by intraperitoneal injections for 2 weeks. Dose, duration and route of pretreatment are based on previous publications^22,48.

Infection models

A detailed protocol for our experimental K. pneumoniae-induced pneumonia model has been previously published⁴⁶. In summary, K. pneumoniae ATCC43816 (K2:O1) was grown from frozen aliquots in Tryptic Soy Broth in a shaking incubator (5% CO2, 37 °C). After overnight growth, one milliliter was transferred to fresh Tryptic Soy Broth, grown to midlogarithmic phase, washed with saline and diluted to a final concentration of 1 × 10⁴ colony-forming units (CFUs) per 50 µL. One day after completion of pretreatment, infection was induced as follows: for the induction of experimental pneumonia, inhalation anesthesia with isoflurane (2-3% in 100% oxygen) was applied and mice were subsequently intranasally inoculated with 50 µL bacterial suspension (10⁴ CFUs). At 12, 24, 36 or 42 h following infection, mice were sacrificed by intraperitoneal injection of ketamine/dexmedetomidine. Of note, for the assessment of ROS release, mice were not infected but sacrificed 24 h after completion of pretreatment.

Following infection and euthanasia, blood was collected via cardiopuncture. The left lung, spleen and liver were harvested and homogenized in four volumes of sterile PBS. Blood and organ homogenates were serially diluted in sterile PBS, plated onto blood agar plates, incubated overnight at 37 °C, and CFUs counted for determination of bacterial growth. Bronchoalveolar lavage fluid (BALF) was collected from right lung lobes using 1 mL phosphate buffered saline (PBS).

Lung pathology scores

Directly after sacrifice, a section of the left lung was fixed in 4% formalin and embedded in paraffin for routine histology. Hematoxylin and eosin-stained paraffin sections from murine lung tissue were scored as previously described^4,50. A specialized pathologist, blinded for pretreatment group, analyzed sections for bronchitis, edema, interstitial inflammation, intraalveolar inflammation, pleuritis, thrombi, and endothelialitis and graded on a scale of 0 (absent) to 3 (severe). The total lung inflammation score was calculated as the sum of the scores for each parameter, the maximum total score being 21.

Flow cytometry

Flow cytometry was performed as earlier detailed^71,72. In brief, BALF was centrifuged at 400 g for 10 min (4 °C). Supernatant was stored at −20 °C until further analysis. The cell pellet was resuspended in sterile PBS, washed and stained following manufacturer’s recommendations with eFluor 780 fixable viability dye (Thermo Fisher), PE-eFluor 610 rat anti-mouse CD45 (clone 30-F11), FITC rat anti-mouse Ly-6G (clone 1A8), and Alexa fluor 647 rat anti-mouse Siglec-F (clone E50-2440; all Biolegend). Following incubation, cells were washed and analyzed by flow cytometry (Cytoflex-S, Beckman Coulter). Precision counting beads (BD Bioscience) were added to quantify cell populations. FlowJo software (Becton Dickinson) was used to detect neutrophils based on surface markers: viable, CD45 + , Siglec F- and Ly-6G+ using FlowJo software (Becton Dickinson), as previously detailed^71,72 and shown in Supplementary Fig. 11.

Murine inflammation and organ damage markers

Inflammatory markers (TNF, IL-6, IL-10, monocyte chemoattractant protein-1) in murine BALF supernatant and plasma were quantified by cytometric bead array (mouse inflammation kit, BD Biosciences) according to the manufacturer’s instructions. IL-1β and chemokine CXC motif ligand (CXCL) 1 were measured in BALF supernatant using a commercial ELISA kit (R&D systems, Minneapolis, MN, US). To quantify oxidative stress in the lungs following pneumonia, nitrotyrosine was measured in lung homogenate by ELISA (Abcam, Cambridge, UK)²¹. The department of clinical chemistry at our hospital (Amsterdam UMC, the Netherlands) quantified organ damage markers (LDH, AST and ALT) in murine blood plasma using a c702 Roche Diagnostics machine.

ROS production assay

ROS production by neutrophils was measured as previously described^50,73. After completion of pretreatment, neutrophils were isolated from murine bone marrow using a single layer of Percoll (62.5%; Sigma). After washing with Hank’s balanced salt solution (HBSS), bone-marrow derived neutrophils were rested in HBSS^+/+ (containing 1.26 mM CaCl₂ and 0.49 mM MgCl₂) for 30 min at room temperature. Neutrophils in HBSS^+/+ were seeded in fetal calf serum precoated 96-well plates at a concentration of 1 × 10⁵ cells/well, together with isoluminol (50 µM; Sigma-Aldrich), and horse radish peroxidase (15 U/mL; Sigma-Aldrich). Neutrophils were stimulated with K. pneumoniae ATCC43816 (K2:O1) (grown as described above) at a concentration of 1 × 10⁶ CFUs/well to achieve a multiplicity of infection of 10. In parallel, phorbol myristate acetate (Sigma) was added as positive control and HBSS^+/+ as negative control. To load bacteria onto cells, plates were centrifuged at 500 × g for 3 min. Immediately after, chemiluminescence was measured every 3 min for 2 h using a Synergy HT plate reader (BioTek).

Statistical analysis

All statistical analyses were performed in R (version 4). Wilcoxon rank-sum tests were used to compare continuous variables between groups. For the EPIC-NORFOLK cohort, competing risk regression models were used to assess associations between the primary outcome (hospitalization or mortality due to lower respiratory tract infections) and tryptophan metabolites, with Benjamini-Hochberg correction for multiple comparisons. Mortality from other causes was treated as competing risk.

To calculate effect sizes for differences in tryptophan metabolites between severe CAP patients and controls, between surviving and non-surviving patients at 90 days, and between participants randomized to controls or antibiotics, we calculated Hedges’ g⁷⁴. For the differences in tryptophan metabolites between CAP patients and controls, linear regression models with log-transformed tryptophan metabolite concentrations were used to control for potential confounding by exposure to any antibiotic prior to sample collection. Since especially anti-anaerobic antibiotics have important effects on gut microbiota, we additionally controlled for exposure to anti-anaerobic antibiotics (piperacillin-tazobactam, meropenem, metronidazole, clindamycin, and amoxicillin with clavulanic acid) using separate linear regression models.

Cox proportional hazards models were used to assess the relationship between log2-transformed IAA concentrations and mortality, at 30 and 90 days, and 1 year. Time zero was defined as the day of ICU admission. Multivariable models were adjusted for age, sex, body mass index, disease severity, antibiotic exposure prior to sample collection, causative pathogen and comorbidities (diabetes, malignancy, immunocompromised state, cardiovascular, renal, and respiratory disease) at ICU admission. We quantified the severity of illness with the Sequential Organ Failure Assessment score. Pearson correlations were calculated between log2-transformed plasma IAA concentrations and the number of ventilator-free days, adjudicated 90 days after ICU admission. The association between IAA at ICU admission and the probability of bacteremia was assessed by univariable logistic regression. For the correlations between tryptophan metabolites and host immune response biomarkers (log2-transformed), Pearson correlations were calculated with Benjamini-Hochberg correction for multiple comparisons. Associations between IAA and gene expression were assessed by linear models, and Gene Set Enrichment Analysis (genes ranked by t-statistic) was applied using Reactome pathways with Benjamini-Hochberg correction^75,76. Reactome pathways related to ‘Immune System’ (R-HSA-168256) were assessed. Linear mixed models were used for dynamics of tryptophan metabolites during experimental pneumonia in mice. Two-tailed level of significance was set at (adjusted) p < 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

R.F.J.K., C.C.A.v.L., B.W.H., B.S., R.v.d.W., A.F.d.V., W.J.d.J., T.v.d.P. and W.J.W. conceived and designed the study. R.F.J.K., C.C.A.v.L., P.S.P. and N.W. performed the mice experiments. T.v.d.P. and O.L.C. were responsible for the MARS cohort. J.J.S. and M.K.B. were responsible for the S3 study. B.W.H., T.S.R.v.E., T.v.d.P. and W.J.W. were responsible for the CAST study. A.L. performed the quantification of tryptophan metabolites. J.J.T.H.R. analyzed the lung sections for pulmonary damage. R.F.J.K., C.C.A.v.L., B.W.H., J.M.B., B.S., R.v.d.W., W.J.d.J., T.v.d.P. and W.J.W. analyzed and interpreted the data. R.F.J.K., B.W.H. and W.J.W. wrote the first draft of the manuscript. All authors have seen and approved the final version of the manuscript.

Peer review

Peer review information

Nature Communications thanks Alessandra Bragonzi, Kurtis Budden, Robert Dickson, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

Source Data are provided with this paper. Transcriptomics data from the MARS cohort have been deposited at the Gene Expression Omnibus public repository of NCBI under accession number GSE65682. Raw mass spectrometry data from MARS, CAST and mice experiments are deposited to PeptideAtlas under dataset identifier PASS05888. Full data of the general-population cohort (EPIC-NORFOLK), severe CAP cohort (MARS) and human intervention trial (CAST) cannot be made publicly available due to privacy issues (i.e. these datasets contain sensitive healthcare information). EPIC-NOFROLK data are available by researchers for specified scientific purposes. Details on how to request data from EPIC-NORFOLK can be found at https://www.epic-norfolk.org.uk/for-researchers/data-sharing/. Source data are provided with this paper.

Code availability

Code used for analyses is publicly available at https://github.com/rfjkullberg/IAA_pneumonia (10.5281/zenodo.15836609)⁷⁷.

Competing interests

WdJ reports grants from the EU and ZonMW, and is a shareholder and board member of AIBiomics BV. WJW reports grants from the EU and ZonMw/NWO, and ad hoc consultancy for AstraZeneca and Shionogi (fees paid to the host institution), outside the submitted work. All other authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-025-63611-y.