Prebiotic activity of lactulose optimizes gut metabolites and prevents systemic infection in liver disease patients

Study Design

This was a prospective cohort study of consecutive hospitalized adult hepatology patients at a single institution from April 2021 to April 2022. Inclusion criteria were age 218 years, ability to provide informed consent (either themselves or by proxy if unable to provide consent), and being treated on the hepatology consult service. Subjects who were younger than 18 years, unable to provide consent, had prior solid organ transplant, or a prior colectomy were excluded. Patients were enrolled as soon as possible upon hospital admission, most within 48 hours. Samples were obtained under a protocol approved by the Institutional Review Board at the University of Chicago (IRB21-0327). Written informed consent was obtained from all participants or their surrogate decision makers.

Specimen Collection and Analysis

Fecal samples were collected as soon as possible following study enrollment. If able to provide additional samples, fecal samples were collected approximately every 2 days during hospitalization. Samples were collected in a similar manner on re-hospitalization up until 1 year post-enrollment or until death or transplant. Demographic and clinical data were collected through review of the medical record and through the University of Chicago Center for Research Informatics (CRI). Clinical laboratory values, death and transplant information was obtained by chart review. Ascitic fluid infections without an evident intraabdominal source were all termed spontaneous bacterial peritonitis (SBP) and were diagnosed with the standard clinical definition of either polymorphonuclear cells (PMN) of 250cells/mm³ or greater or a positive ascites culture^43,44. Cultures growing common contaminants (i.e. components of skin flora) were considered negative if PMN were < 250cells/mm³ and the clinical team did not treat for SBP. Samples were paired with the closest stool sample that was within 14 days prior to the ascitic sample or 3 days after the ascitic sample. If a sample was positive, all subsequent samples for the next 30 days were considered surveillance paracenteses and excluded from analysis. If a sample was negative, subsequent samples, which were typically done to for volume management, were also excluded. This approach mitigated observing the effects of targeted antibiotic use on the microbiome after diagnosis of SBP, over-representing a single patient for repeated negative paracenteses, and ensured that each analyzed stool sample was paired with only one ascites sample. Blood stream infection was defined as having a blood culture with bacterial or fungal growth. Skin contaminants were again excluded (i.e. considered “negative” cultures) if the clinical team did not treat for bacteremia. Blood cultures were paired to the closest stool sample within 30 days prior or 7 days after the blood culture. Blood cultures 30 days after a positive culture (i.e. “surveillance cultures”) were excluded from analysis, and blood cultures drawn within 30 days after a negative culture were also excluded as “duplicates.”

Metagenomic Analyses

Fecal samples underwent shotgun DNA sequencing. After undergoing mechanical disruptions with a bead beater (BioSpec Product), samples were further purified with QIAamp mini spin columns (Qiagen). Purified DNA was quantified with a Qubit 2.0 fluorometer and sequenced on the Illumina HiSeq platform, producing around 7 to 8 million PE reads per sample with read length of 150 bp. Adapters were trimmed off from the raw reads, and their quality were assessed and controlled using Trimmomatic (v.0.39)⁵⁷, then human genome were removed by kneaddata (v0.7.10, https://github.com/biobakery/kneaddata). Taxonomy was profiled using metaphlan4 using the resultant high-quality reads⁵⁸. Microbial reads then were assembled using MEGAHIT (v1.2.9)⁵⁹, genes are called by prodigal (https://github.com/hyattpd/Prodigal). In addition, high-quality reads are queried against genes of interest, such as virulence factors, cazymes, and antibiotic resistance genes, using DIAMOND (v2.0.4)⁶⁰, and hits are filtered with threshold > 80% identity, > 80% protein coverage, then abundance is tabulated into counts per million or reads per million mapped reads (RPKM).

Alpha-diversity of fecal samples was estimated using the Inverse Simpson Index, while beta diversity was assessed by using taxumap (https://github.com/jsevo/taxumap). Metagenomic information is publicly available on NCBI under BioProject ID PRJNA912122.

Mouse fecal DNA isolation

DNA was extracted using the QIAamp PowerFecal Pro DNA kit (Qiagen). Prior to extraction, samples were subjected to mechanical disruption using a bead beating method. Briefly, samples were suspended in a bead tube (Qiagen) along with lysis buffer and loaded on a bead mill homogenizer (Fisherbrand). Samples were then centrifuged, and supernatant was resuspended in a reagent that effectively removed inhibitors. DNA was then purified routinely using a spin column filter membrane and quantified using Qubit.

16S sequencing

V4-V5 region within 16S rRNA gene was amplified using universal bacterial primers – 563F (5’-nnnnnnnn-NNNNNNNNNNNN-AYTGGGYDTAAA-GNG-3’) and 926R (5’-nnnnnnnn-NNNNNNNNNNNN-CCGTCAATTYHT-TTRAGT-3’), where ‘N’ represents the barcodes, ‘n’ are additional nucleotides added to offset primer sequencing. Approximately ∼412bp region amplicons were then purified using a spin column-based method (Minelute,Qiagen), quantified, and pooled at equimolar concentrations. Illumina sequencing-compatible Unique Dual Index (UDI) adapters were ligated onto the pools using the QIAseq 1-step amplicon library kit (Qiagen). Library QC was performed using Qubit and Tapestation and sequenced on Illumina MiSeq platform to generate 2×250bp reads.

16S qPCR

Extracted DNA was diluted to 20ng/ul to ensure concentrations fell within measurable range. Degenerate primers were diluted to 5.5mM concentration. Primer sequences are as follows: 563F (5’-AYTGGGYDTAAAGNG-3’) and 926Rb (5’-CCGTCAATTYHTTTRAGT-3’). Standard curves were generates using linearized TOPO pcr2.1TA vector (containing V4-V5 region of the 16S rRNA gene) transformed into DH5α competent bacterial cells. Five-fold serial dilution was performed on the purified plasmid from 10⁸ to 10³ copies/□l per tube. PCR products were detected using PowerTrack SYBR Green Master Mix (A46109). qPCR was run on QuantStudio 6 Pro (Applied Biosystems) with the following cycling conditions: 95°C for 10 min, followed by 40 cycles of 95°C for 30 s, 52°C for 30s, and 72°C for 1min. Copy numbers for samples were calculated using the Design and Analysis v2 software

Metabolomic Analyses

Short chain fatty acids (SCFA, i.e. butyrate, acetate, propionate, and succinate) were derivatized with pentafluorobenzyl bromide (PFBBr) and analyzed via negative ion collision induced-gas chromatography-mass spectrometry ([–]CI-GC-MS, Agilent 8890)⁶¹. Eight bile acids (BA, i.e. primary: cholic acid; conjugated primary: glycocholic acid, taurocholic acid; secondary: deoxycholic acid, lithocholic acid [LCA], isodeoxycholic acid; modified secondary: alloisolithocholic acid [alloisoLCA] and 3-oxolithocholic acid [3-oxoLCA]) were quantified (µg/mL) by negative mode liquid chromatography-electrospray ionization-quadrupole time-of-flight-MS ([–]LC-ESI-QTOF-MS, Agilent 6546). Eleven indole metabolites were quantified by UPLC-QqQ LC-MS. Eighty-five additional compounds were relatively quantified using normalized peak areas relative to internal standards. Data analysis was performed using MassHunter Quantitative Analysis software (version B.10, Agilent Technologies) and confirmed by comparison to authentic standards. Quantitative fecal metabolomic information paired to fecal metagenomic information is publicly available on NCBI under BioProject ID PRJNA912122. Raw data files are publicly available on MetaboLights project ID MTBLS7046.

Bacterial culture

The Bifidobacteria longum strains MSK.11.12 and DFI.2.45 were previously derived from two distinct healthy donor stool samples and whole genome sequenced (BioSample ID: SAMN19731851 and SAMN22167409). The vancomycin resistant Enterococcus faecium (VRE) strain used in this study was obtained from ATCC (strain 700221). Both bacterial strains were grown in anaerobic conditions in Brain-heart infusion broth (BHI broth, BD 237500). The pH was adjusted to 7.0 with NaOH. Media was supplemented with lactulose (Thermo Scientific, J60160-22) or sucrose (Fisher BP220-1) where indicated. For growth in media supplemented with short chain fatty acids, media was supplemented with either sodium butyrate (Sigma 303410), sodium succinate (Sigma S2378) or sodium acetate (Sigma S5636). For studies of bile acid metabolism, 10μg/mL of conjugated primary bile acid, either glycocholic acid (Sigma T4009) or taurocholic acid (EMD Millipore, 360512), was added to the media. All growth curves were obtained in anaerobic culture conditions at 37°C on a BioTek EPOCH2 microplate reader with BioTek Gen 5 3.11 software. Growth curves were analyzed in GraphPad Prism Version 9.4.0. Lactulose concentrations in were measured using the EnzyChrom™ Lactulose Assay Kit (ELTL-100).

Mouse studies

All mouse studies were approved by The University of Chicago Institutional Animal Care and Use Committee (IACUC, Protocol 72599). For germ-free studies, 8-12 week-old male C57BL/6 mice were used. Mice were bred and raised in a germ-free isolator, and after removal from the germ-free isolator, mice were handled in a sterile manner and individually housed in sealed negative pressure (BCU) isolators. Mice were treated with either regular sterile water or sterile water supplemented with filter sterilized lactulose at a final concentration of 20g lactulose/L of water. B. longum was grown to steady state in BHIS, pelleted, and resuspended in an equal volume of PBS, and previously germ-free mice were gavaged with 200μL of a freshly prepared suspension on 3 consecutive days. Fecal pellets were collected prior to and after B. longum inoculation at the indicated timepoints for 16S rRNA metagenomic analysis and targeted metabolomic analysis.

Data sharing

In addition to the repositories specified above, all raw data included in the manuscript is publicly available at: https://github.com/yingeddi2008/DFILiverDiseaseMicrobiome.

Patient recruitment and disease characteristics

We enrolled 356 hospitalized patients with liver disease from April 2021 through April 2022. Of these 356 patients, 263 (73.8%) produced 850 stool samples that were analyzed by shotgun metagenomics and targeted metabolomics. The demographics and admission disease characteristics of patients who produced samples are shown in Table 1. Notably, alcohol use was the most common etiology of liver disease and accounted for 51.0% of patients. Patients had a median model of end stage liver disease MELD-sodium (MELD-Na) score of 18.69 (IQR 10.61 – 27.88). Of the 263 patients included in the analysis, 183 (69.6%) had known clinically significant portal hypertension upon hospitalization, which was defined either by hepatic venous pressure gradient (HVPG) ≥ 10mmHg, characteristic findings on imaging (enlarged portal vein, intraabdominal varices, splenomegaly, or ascites), or clinical features of ascites, varices, or hepatic encephalopathy. Over half (137, 52.1%) of the enrolled patients had some form of end organ dysfunction present on admission. End organ dysfunction included neurologic dysfunction (grade III or higher HE), circulatory dysfunction (mean arterial pressure < 60mmHg or requiring vasoactive agents), respiratory dysfunction (new or increasing oxygen requirement), or renal dysfunction (acute kidney injury or new hemodialysis needs). Taken together, these enrollment demographics and baseline disease characteristics indicate that, on average, this patient cohort has advanced liver disease with a poor prognosis.

Fecal microbiome and metabolite profiles of liver disease patients are highly variable

To determine the range of fecal microbiome compositions in patients with liver disease, we performed shotgun sequencing on DNA extracted from 850 fecal samples collected from 263 liver disease patients. We used MetaPhlan4 to assign taxonomic compositions and Bray-Curtis distance to assess microbial diversity within the resulting dataset (Figure 1A). Immediately apparent is the low taxonomic alpha diversity in the majority of fecal samples (inverse Simpson range: 1.00 – 36.41, mean = 6.22, median = 3.86) and the expansion of bacterial taxa belonging to the Enterococcus genus (forest green) and Enterobacteriaceae family (red), which are common hallmarks of dysbiosis and are similar to microbiome domination detected in patients undergoing allogeneic hematopoietic cell transplantation^24,62. Another striking observation is the high number of samples with marked expansion of the genus Bifidobacteria (shown in purple), which is commonly reported in breast fed infants but rarely seen in adults, suggesting that this may be a liver disease-specific finding.

• Download figure
• Open in new tab

Figure 1: Fecal samples from hospitalized patients with liver disease display a wide range of microbiome compositions and metabolomic profiles.

Hierarchical clustering of all 850 fecal samples is based on bray-curtis distance binned into nine clusters. (A) Relative taxa abundance by shotgun metagenomics is shown for each sample in the top panel, and (B and C) quantitative targeted metabolites for each corresponding sample are shown below. There is a wide variation in metagenomic compositions with some samples showing high biodiversity with anaerobic commensal bacteria compared to others that were dominated by a single taxon. (B) Short chain fatty acids and (C) both secondary and modified secondary bile acids are generally more abundant in the more biodiverse specimens. Units for SCFA are mM, and units for BA derivatives are in μg/mL.

While microbiome-derived or modified metabolites mediate many of the beneficial impacts on mucosal immune defenses and epithelial barrier functions, little is known about their production in liver disease patients harboring vastly different microbial populations. We used targeted GC- and LC-MS platforms to determine relative amounts 82 metabolites in all 850 fecal samples within our dataset, and the first sample from each patient is shown (Figure S1). Ranking of fecal samples from lowest to highest alpha diversity demonstrates the correlation between microbiome diversity and the relative amounts of SCFAs, branched chain FAs, aminated FAs, secondary bile acids, and various indole compounds and tryptophan metabolites. As expected, low diversity fecal samples have markedly reduced amounts of secondary bile acids but also reduced concentrations of SCFAs and a subset of indole compounds.

In order to obtain a more quantitative picture of fecal metabolite concentrations, we measured the concentration of SCFAs and primary and secondary bile acids in fecal samples from patients with liver disease. Despite some variation, higher proportions of commensal anaerobes and alpha-diversity generally coincided with greater levels of all measured short chain fatty acids (towards the right side of Figure 1B). Additionally, samples with more commensal anaerobes tended to have greater capacity to deconjugate primary bile acids and generate both secondary and modified secondary bile acids (Figure 1C). These findings demonstrate that the fecal microbiome and metabolome is highly variable in hospitalized patients with liver disease.

Fecal metabolites correlate with expansion of distinct microbial taxa

To more precisely associate fecal metabolite concentrations with fecal microbiome compositions, fecal samples were assigned to 1 of 12 taxonomic groups on a taxonomic Uniform Manifold Approximation and Projection (taxUMAP) (Fig 2A). While all clusters displayed a wide range of short chain fatty acid production, our analysis presented in Figure 2 revealed that clusters that were characterized by samples with Enterococcus (forest green), Proteobacteria (red), or Lactobacillus (royal blue) as the most abundant taxa all had lower levels of short chain fatty acid production when compared to samples characterized by expansion of Bifidobacteria (purple), Lachnospiraceae (pink) or Bacteroidetes (aquamarine) (Fig 2A-D, S2A-C). On average, Bifidobacteria-expanded samples, which are prevalent in patients with liver disease, had higher acetate production (Fig 2B, S2A) but similar levels of butyrate (Fig 2C, S2B) and propionate (Fig 2D, S2C) compared to samples with high Bacteroidetes or Lachnospiraceae abundance. Consistent with a role for Bifidobacteria in acetate production, samples with greater than 10% Bifidobacteria abundance had a positive and statistically significant linear correlation with increasing acetate but not butyrate (Fig S2A-B, right-most panels). Propionate concentration negatively correlated with Bifidobacteria abundance (Fig S2C, right-most panel).

• Download figure
• Open in new tab

Figure 2: The unique Bifidobacteria expanded cluster has a distinct metabolite profile.

(A) A taxUMAP was generated using 850 samples from 263 patients with liver disease. Each sample is represented by a single dot that is colored based on the most abundant taxon as indicated. Samples within the taxUMAP were pseudocolored based on the indicated (B-D) short chain fatty acid (SCFA) or (E-G) bile acid concentrations. Units for SCFA are mM, and units for BA derivatives are in μg/mL. Samples dominated by potentially pathogenic taxa (i.e. Enterococcus and Proteobacteria) were deficient in SCFA generation and generation of secondary bile acids compared to those dominated by anaerobic species. The unique Bifidobacteria cluster was particularly adept at generating acetate but produced secondary bile acids less efficiently than other anaerobe-dominated samples.

While all groups had detectable concentrations of primary bile acids, Enterococcus dominated samples had the highest (Fig S3). In contrast, conjugated primary bile acid levels were lower (Fig 2E, S3) and secondary (Fig 2F, S3) and modified secondary BA (Fig 2G, S3) levels were higher in samples with high levels of obligate anaerobes, including Bifidobacteria, Bacteroidetes, and Lachnospiraceae. Samples dominated by Bifidobacteria had higher levels of primary bile acids and lower levels of secondary bile acids compared to Bacteroidetes- and Lachnospiraceae-dominated samples, suggesting Bifidobacteria provide increased bile acid hydrolase activity and/or reduced efficiency of primary BA to secondary BA conversion. Taken together, our results indicate that a subset of patients with liver disease are predisposed to Bifidobacteria expansion with resultant acetate production and conjugated bile salt hydrolysis with reduced efficiency of secondary bile acid generation.

Lactulose administration is associated with increased Bifidobacteria abundance and reduced antibiotic-resistant pathogen density

As lactulose is a common therapy administered to liver disease patients and has been reported to act as a prebiotic, we next sought to address how it might be impacting microbiome composition. Fecal samples obtained from liver disease patients after lactulose administration (lactulose given within 7 days prior to fecal sample) were taxonomically diverse, with many having marked Bifidobacteria expansion. In many samples, Bifidobacteria were the predominant genus (Fig 3A, B; range: 0.0 – 94.7%, mean: 10.9%, stdev: 10.9%). This contrasted with fecal samples obtained from patients who were not treated with lactulose (Fig 3A, B; range: 0.0 – 50.1%, mean: 3.57%, stdev: 3.57%, p < 0.001). High abundances of potentially pathogenic bacteria including Enterobacteriaceae and Enterococcus were detected in lactulose treated and untreated patients (Fig 3A). Enterococcus expansion was largely restricted to low-Bifidobacteria abundance samples and was associated with the presence of the vancomycin-resistance gene vanA, a result that was more striking in samples obtained in the presence of lactulose (Fig 3C). These results suggest that lactulose-mediated Bifidobacteria expansion may limit the expansion of Vancomycin-resistant Enterococcus (VRE) in the gut. Alternatively, treatment with broad-spectrum antibiotics may reduce the density of Bifidobacterium species while also eliminating other commensal species mediating resistance to VRE in lactulose-treated patients (Fig 3B) ^24,63. Although the osmotic laxative effect of lactulose may affect the fecal metagenomic composition, we were unable to detect an association between stool consistency and Bifidobacteria abundance (Supplemental Data).

• Download figure
• Open in new tab

Figure 3: Lactulose use is associated with increased Bifidobacteria species abundance and reduced VRE abundance in the absence of systemic antibiotic use.

(A) Relative abundance by shotgun taxonomy of all 850 samples is shown. Samples were arranged first by whether they were obtained without lactulose (left facet) or after lactulose administration (right facet) and then arranged by decreasing Bifidobacterium species (shown in purple) relative abundance from left to right. With decreasing Bifidobacteria abundance, there is higher abundance of Enterococcus (shown in forest green). (B) Medication information regarding lactulose, rifaximin, and broad-spectrum antibiotics is shown. Red indicates that a given treatment was administered prior to the sample being collected, and green indicates that a medication was not given prior to the sample being collected. (C) Relevant antibiotic resistance (vanA and vanB) genes were queried for each sample and are pseudocolored based on expression level normalized to gene length (RPKM).

Lactulose-mediated Bifidobacteria expansion is associated with increased levels of multiple bioactive fecal metabolites

Bifidobacteria expansion, defined here as relative abundance ≥10%, was associated with marked changes in fecal metabolite profiles only in samples obtained in the presence of lactulose (Fig 4, S4). In lactulose-treated patients with Bifidobacteria expansion, the fecal metabolome was characterized by increased SCFA concentrations (including acetate, butyrate, and propionate), reduced conjugated primary bile acid concentrations (taurocholic, glycocholic, taurochenodeoxycholic, and glycochenodeoxycholic acids), increased secondary (lithocholic, isodeoxycholic acid) and modified secondary (alloisolithocholic, 3-oxo-lithocholic acid) bile acids, and indole metabolites (indole-3-lactic acid and indole-3-carboxaldehyde) (Fig 4A-C). Consistent with a known property of Bifidobacteria to hydrolyze conjugated primary bile acids, ratios of primary bile acids to conjugated primary bile acids were substantially higher in lactulose-treated samples with Bifidobacteria expansion (Fig 4D). While secondary bile acid levels were also higher in these samples, this was primarily due to an increased capacity to deconjugate primary bile acids rather than convert primary BA to secondary BA (Fig 4D, middle and right-most panels). In the absence of lactulose treatment, fecal samples with greater than 10% Bifidobacteria also had increased concentrations of SCFA (acetate and butyrate) and decreased conjugated primary bile acids (Fig S4). However, these changes were all modest compared to those from patients receiving lactulose, suggesting that lactulose impacts the metabolic activities of the gut microbiome in patients with liver disease (Fig S4).

• Download figure
• Open in new tab

Figure 4: Lactulose-mediated Bifidobacteria expansion is associated with significant changes in bioactive fecal metabolites.

(A) Volcano plot (log₂fold change vs. log₁₀p-value) of qualitative metabolites comparing samples with low (< 10%) vs. high (≥10%) Bifidobacteria abundance after lactulose exposure. P-values are corrected for multiple comparisons. Values with log2 fold-change > 1 (corresponding to a 2-fold change with a p-value < 0.05) were considered significant. (A and B) Notably, low Bifidobacteria abundance is associated with higher levels of unconjugated primary bile acids, whereas high Bifidobacteria abundance is associated with higher levels of beneficial bacterial metabolites, including short chain fatty acids, secondary and modified secondary bile acids, and immunomodulatory indole metabolites. (C) Bar graphs for select SCFA (acetate, butyrate, and propionate) and bile acids (taurocholic, lithocholic, and alloisolithocholic acid) where each point represents a single value. Median and interquartile range are indicated by the line and box, respectively. Statistical comparisons between groups were analyzed using the Wilcoxin rank sum. Units for SCFA are mM, and units for BA derivatives are in μg/mL. (D) Bile acid conversion from conjugated-primary BA to primary BA and then to secondary BAs was tested for each sample. Each point represents a ratio for an individual sample. Samples were grouped by whether they had expanded Bifidobacteria in the presence of lactulose. CA: cholic acid; GCA: glycocholic acid; DCA: deoxycholic acid.

Lactulose enhances Bifidobacteria longum growth, acetate production and inhibition of VRE

Many fecal samples obtained from lactulose treated patients had high abundances of potentially pathogenic bacteria belonging to Enterobacteriaceae family (range: 0.0 – 99.2%, mean: 8.04%, median: 0.73%, stdev: 8.04%) and the Enterococcus genus (range: 0.0 – 100%, mean: 30.8%, median: 5.24%, stdev: 30.8%), with, on average, higher densities than those seen in fecal samples taken from patients not receiving lactulose (Proteobacteria mean: 6.38%, median: 1.25%; Enterococcus mean: 7.4%, median: 0.06%). While high abundances of Proteobacteria species were seen throughout a range of Bifidobacteria abundances, Enterococcus domination was seen predominantly in samples with reduced to absent Bifidobacteria abundance (Fig 3A, 5A).

• Download figure
• Open in new tab

Figure 5: Lactulose-mediated Bifidobacteria expansion is associated with exclusion of antibiotic-resistant Enterococcus species.

(A) Relative abundance of potentially pathogenic taxa Enterococcus and Proteobacteria were plotted based on lactulose exposure (no lactulose, left; lactulose, right) and relative abundance of Bifidobacteria (< 10%, blue; ≥10%, red). Neither taxa was changed at high or low Bifidobacteria abundance without lactulose, but with lactulose, Enterococcus abundance was significantly increased with low Bifidobacteria abundance in human fecal samples. Proteobacteria are modestly but statistically significantly increased in Bifidobacteria expanded samples from lactulose-treated patients. (B) A human-derived B. longum species was grown in BHIS media (blue) or BHIS supplemented with 50mM lactulose (red) or 50mM sucrose (green), and growth curves (OD₆₀₀ over time) are shown. Growth curves are representative of three independent experiments done in at least triplicate. B. longum preferentially uses lactulose as an energy source. (C) Schematic of experimental design for B. longum and VRE co-culture. B. longum and VRE were grown to steady state and diluted to low density (OD₆₀₀ = 0.05) prior to inoculation in either BHIS or BHIS supplemented with 50mM lactulose. Cultures were inoculated with both bacteria simultaneously (top) or B. longum was given either a 24-hour or 48-hour lead time prior to VRE inoculation (bottom). (D) After 24-hours of co-culture, serial dilutions were plated for VRE c.f.u. counts. Groups were compared using a two-tailed student’s t-test. The plot is from one experiment that is representative of three independent experiments done in at least triplicate.

To further investigate the inverse relationship between Bifidobacteria and VRE abundance, we isolated a Bifidobacterium longum strain from a healthy human donor and performed whole genome sequencing to demonstrate the presence of genes involved in lactulose metabolism and acetate production^64,65. B. longum growth in culture was markedly augmented by addition of lactulose to media, and to a lesser extent by addition of sucrose (Fig 5B). Lactulose-induced growth augmentation was accompanied by increased media acidification and acetate, but not butyrate or propionate, production (Fig S5A, B). Similar to patient fecal samples dominated by Bifidobacteria and consistent with coding for bile salt hydrolase genes, monocultures of B. longum efficiently deconjugated conjugated primary bile acids but did not convert these to secondary bile acids (Fig S5C, D).

We next colonized germ-free (GF) mice with B. longum in the presence or absence of lactulose (Fig S6A). Lactulose was detectable in the stool of GF mice and led to increase stool water content (Fig S6B, C). B. longum efficiently colonized and persisted in the intestines of gnotobiotic mice independent of lactulose (Fig S6D). Consistent with in vitro functions, B. longum-colonized mice had measurable fecal acetate levels but did not generate butyrate or propionate (Fig S6E-G). Fecal acetate levels were similar between water and lactulose-treated mice at day 0, but on day 10 lactulose augmented acetate production (Fig S6E). On day 10 after B. longum colonization, fecal bile acid profiling demonstrated increased primary bile acids but undetectable secondary BA (Fig S6H-J).

To test whether B. longum inhibits VRE growth and to determine the impact of lactulose administration, we performed co-culture assays (Fig 5, C-D). While simultaneous inoculation of B. longum and VRE into media did not reduce VRE growth (Fig 5C-D), pre-culture of B. longum for 24h or 48h fully inhibited VRE growth in the presence of lactulose and only partially inhibited VRE growth in the absence of lactulose (Fig 5C-D). Serial dilutions of B. longum culture supernatant with fresh media sequentially reduced the inhibitory effect on VRE growth (Fig S7A, B), and alkalization of B. longum supernatant pH from 5.8 up to 7.2 restored VRE’s initial log phase growth (Fig S7C) but did not enable VRE to reach the maximal OD₆₀₀ obtained in fresh, neutral pH media (Fig S7C). VRE growth is inhibited by addition of short chain fatty acids to the media, and inhibition is augmented at lower media pH (Fig S7D). Taken together, these data suggest that B. longum inhibits VRE growth through parallel mechanisms that include media acidification, SCFA production, and, to a lesser extent, nutrient deprivation. B. longum-mediated VRE inhibition is significantly augmented by lactulose, potentially explaining the expansion of VRE seen in fecal samples lacking expanded Bifidobacteria populations despite lactulose administration. Consistent with VRE expansion seen in patients treated with lactulose without Bifidobacteria expansion, VRE growth is also augmented by lactulose (Fig S7E,F).

Bifidobacteria expansion and associated metabolite production correlates with reduced incidence of infections in advanced liver disease

To determine whether lactulose-mediated Bifidobacteria expansion and associated metabolite changes are associated with clinical benefits, we correlated the gut microbiome metagenomic and metabolic data with development of common infections in cirrhosis, including spontaneous bacterial peritonitis (SBP) and bacteremia (Figs 6, S8, and S9).

• Download figure
• Open in new tab

Figure 6: Bifidobacteria expansion and associated metabolite production are associated with decreased incidence of spontaneous bacterial peritonitis (SBP) and bacteremia and prolonged survival.

(A) A total of 19 episodes of SBP were identified that had an adjacent stool sample. A Linear discriminant analysis effect size (LEfSe) showing the significant (Wilcoxon rank-sum, two-tailed, p ≤ 0.05) effect sizes of taxa between groups (purple bars: no SBP, green bars: SBP). (B) Odds ratio of being diagnosed with SBP if Bifidobacteria relative abundance is greater than 10% was calculated from the data shown. (C) When compared to fecal samples from patients with SBP, those without SBP had statistically (Wilcoxon rank-sum, two-tailed, p < 0.05) higher concentrations of acetate, and lower concentrations of the conjugated primary bile acid taurocholic acid. (D) A total of 23 episodes of bacteremia were identified from 228 unique blood cultures that had a stool sample within the pre-defined time window relative to blood culture. A Linear discriminant analysis effect size (LEfSe) showing the significant (Wilcoxon rank-sum, two-tailed, p ≤ 0.05) effect sizes of taxa between groups (yellow bars: no bacteremia, blue bars: bacteremia). (E) Odds ratio of being diagnosed with bacteremia if Bifidobacteria relative abundance is greater than 10% was calculated from the data shown. (F) When compared to fecal samples from patients with bacteremia, those without bacteremia had statistically (Wilcoxon rank-sum, two-tailed, p < 0.05) similar levels of short chain fatty acids and lower concentrations of the conjugated primary bile acid taurocholic acid. Units for SCFA are mM, and units for BA derivatives are in μg/mL. (G) Kaplan-Meier curves were generated for survival based on initial stool sample characteristics. Survival curves were stratified based on lactulose administration and Bifidobacteria expansion of the initial stool sample. The number at risk at each time-point is shown below.

We identified 111 ascites samples with near concurrent fecal samples for analysis, among which there were 19 diagnoses of SBP (Fig 6A, B, S8A, Table S1). LEfSe analysis demonstrated that Bifidobacteria was strongly associated with resistance to SBP, while presence of the mucus-degrading Akkermansia was associated with SBP (Fig 6A). We used a frequency cutoff of 10% to distinguish patients with high Bifidobacteria from those with low Bifidobacteria frequencies. We found that 1 of 29 patients (3.4%) with ≥10% Bifidobacteria frequency was diagnosed with SBP while 18 of 63 patients (28.6%) with less than 10% Bifidobacteria developed SBP (Fig 6B; Odds Ratio 0.12, 95% CI 0.01 – 0.79, p value 0.021). Qualitative metabolomics demonstrated that glycholithocholic acid and serotonin were significantly associated with being SBP-free (Fig S8B). The Bifidobacteria-generated SCFA acetate was also relatively increased in SBP-free samples, albeit to a lesser extent (p < 0.05, log₂ fold-change 0.9, Fig S8B). Quantitative metabolomics demonstrated that acetate was significantly higher in SBP-free fecal samples and the conjugated primary bile acid taurocholic acid was significantly higher in samples with SBP (Fig 6C). Consistent with a role for Bifidobacteria in protection from SBP, samples from patients without SBP had a significantly greater conversion of conjugated to unconjugated primary BA (CA:GCA ratio) but similar conversion of primary to secondary BA (Fig S8D).

To correlate fecal microbiome compositions and metabolite concentrations with development of bacteremia, we paired 328 blood cultures with fecal samples for analysis. Blood culturing identified 23 distinct diagnoses of bacteremia (Fig 6D, E, S9A, Table S2) and LEfSe analysis demonstrated that Bifidobacteria was associated with remaining bacteremia-free (Fig 6D). There were 2 diagnoses of bacteremia in 57 patients (3.5%) with ≥10% Bifidobacteria and 21 of 148 patients (14.2%) with <10% Bifidobacteria (Fig 6E; Odds Ratio 0.247, 95% CI 0.056 – 0.996, p value 0.047). Qualitative metabolomics demonstrated that conjugated primary BA (glycocholic, taurocholic, and glycochenodeoxycholic acid) and the primary BA cholic acid were all significantly increased in bacteremia-associated samples (Fig S9B). Targeted quantitative metabolomics confirmed that samples from bacteremic patients had reduced capacity to hydrolyze conjugated bile acids (Fig 6F, S9D). These data suggest that lactulose-mediated Bifidobacteria expansion with resultant SCFA production and bile salt hydrolase activity may play a protective role against some of the most common and devastating infectious complications of advanced liver disease.

Lactulose-mediated gut Bifidobacteria expansion is associated with improved transplant-free survival

Finally, we determined whether initial stool characteristics can identify patients at increased risk of adverse outcomes and thus might serve as targets for microbiome-based therapies. To do so, we assessed 90-day overall survival in patients stratified by lactulose exposure and Bifidobacteria expansion. Compared to patients who received lactulose but did not expand Bifidobacteria, patients with lactulose-associated Bifidobacteria expansion had significantly improved overall 90-day survival (Fig 6G) with similar rates of liver transplantation (Fig S10). While many patient characteristics between the groups were similar (including age, race, BMI, cause of liver disease, and presence of clinically significant portal hypertension), patients who received lactulose but did not expand Bifidobacteria had clinical parameters that correlate with a poor prognosis (including higher MELD-Na scores and higher incidence of additional organ failures present upon hospitalization) (Table 3). This suggests that Bifidobacteria expansion and associated metabolite profiles can serve as a biomarker for healthier patient cohort, which may be due to the gut microbiome differences along with other disease-specific factors.