Hyperimmune bovine colostrum containing lipopolysaccharide antibodies (IMM124-E) has a nondetrimental effect on gut microbial communities in unchallenged mice

ABSTRACT

Enterotoxigenic Escherichia coli (ETEC) is a leading cause of bacterial diarrhea with the potential to cause long-term gastrointestinal (GI) dysfunction. Preventative treatments for ETEC-induced diarrhea exist, yet the effects of these treatments on GI commensals in healthy individuals are unclear. Whether administration of a prophylactic preventative treatment for ETEC-induced diarrhea causes specific shifts in gut microbial populations in controlled environments is also unknown. Here, we studied the effects of a hyperimmune bovine colostrum (IMM-124E) used in the manufacture of Travelan (AUST L 106709) on GI bacteria in healthy C57BL/6 mice. Using next-generation sequencing, we aimed to test the onset and magnitude of potential changes to the mouse gut microbiome in response to the antidiarrheagenic hyperimmune bovine colostrum product, rich in immunoglobulins against select ETEC strains (Travelan, Immuron Ltd). We show that in mice administered colostrum containing lipopolysaccharide (LPS) antibodies, there was an increased abundance of potentially gut-beneficial bacteria, such as Akkermansia and Desulfovibrio, without disrupting the underlying ecology of the GI tract. Compared to controls, there was no difference in overall weight gain, body or cecal weights, or small intestine length following LPS antibody colostrum supplementation. Overall, dietary supplementation with colostrum containing LPS antibodies produced subtle alterations in the gut bacterial composition of mice. Primarily, Travelan LPS antibody treatment decreased the ratio of Firmicutes/Bacteroidetes in gut microbial populations in unchallenged healthy mice. Further studies are required to examine the effect of Travelan LPS antibody treatment to engineer the microbiome in a diseased state and during recovery.

INTRODUCTION

Gut bacteria are part of a diverse population of commensal microorganisms that modulate host behavior and physiology. Endogenous bacteria influence the development of the gastrointestinal (GI) tract and immune system, the production of vitamins and release of neurotransmitters that regulate intestinal motility and fluid secretion. Microbes also assist the metabolism of otherwise indigestible dietary carbohydrates. Bacterial flora influence the structure and function of the GI mucosal barrier and shape the broader physiology of the host, including impacts on the central nervous system and behavior via gut-brain axis pathways.

Enterotoxigenic Escherichia coli (ETEC) is a major cause of bacterial diarrhea in children in developing countries, military personnel, and travelers (1 – 4). ETEC-induced diarrhea (also known as Traveler’s Diarrhea) is usually self-limiting, however, it can contribute to impairments in developmental processes or even mortality in young children (5) and long-term post-infection effects such as irritable bowel syndrome in travelers (6). Alterations in gut microbiome profiles in response to infections with gram-negative pathogens responsible for Traveler’s Diarrhea have been identified (4, 7), yet the effects of ETEC antibodies on endogenous bacteria remain unclear.

Lipopolysaccharide (LPS) is a major component of the outer membrane of gram-negative bacteria such as E. coli. LPS is composed of a lipid A endotoxin connected via an oligosaccharide core to a hypervariable O-antigen polysaccharide (8). The lipid A component is a potent activator of innate immune signaling via the Toll-like receptor 4/Myeloid differentiation factor 2 (TLR4/MD-2) complex (9). At high concentrations, lipid A induces acute inflammation resulting in fever, tachycardia, and septic shock in mammals, including humans (as reviewed in references 10 and 11). Orally ingested bovine immunoglobulins are an emergent therapeutic under investigation as both preventative (12) and potential treatment of GI dysfunction resulting from LPS-induced inflammation (reviewed by references 13 and 14). Immunoglobulins targeting LPS can bind pathogens to prevent entry to host tissue via the mucosal epithelium. Furthermore, antibodies against LPS can modulate immune responses to enhance pathogen clearance and improve metabolic function by interacting with microorganisms in the gut (reviewed by references 14 – 18). Although LPS influences host physiology via the microbiome (19, 20), whether LPS antibody supplementation alters microbial populations is not known. Here, we investigated the effects of dietary supplementation of colostrum-containing LPS antibodies on GI microbial richness and diversity in mice.

MATERIALS AND METHODS

Preparation of colostrum and E. coli antibodies

Holstein Friesian and Jersey dairy cows at commercial dairy farms were immunized with either monovalent or polyvalent ETEC vaccines containing 0.5 mg of purified proteins administered subcutaneously in three 1 mL doses. The vaccines contain three important pathogenic and antigenic determinants; LPS, flagella, and colonization factor antigen (CFA), which collectively play roles in bacterial membrane stability, immune evasion, motility, and adherence (21, 22). Three ETEC doses were administered prior to calving at 9–12 weeks, 6 weeks, and 1 month before calving. The monovalent vaccine used to immunize each animal contained the common ETEC strain; serotype O78, and the polyvalent vaccine contained a combination of serotypes (O6, O8, O15, O25, O27, O63, O114, O115, O128, O148, O153, and O159) that were selected to cover the majority of E. coli that cause Traveler’s Diarrhea (Table 1). To produce a rapid and durable immune response, Montanide ISA 206 veterinary adjuvant (approved by the Australian Pesticides and Veterinary Medicines Authority) was used in the preparation of the vaccine. Inoculation with these outer antigens activates a generalized immune response in the host animal to produce antibodies (mainly IgG) which recognize and bind with the bacterial cell-surface epitopes presented. These polyclonal antibodies have been shown to cross-react with the LPS from a wide range of ETEC O serotypes, as well as Shigella and Salmonella O polysaccharides (15). The Immuron Ltd hyperimmune bovine colostrum (IMM-124E) is a pasteurized, fat, and lactose-reduced spray dried powder that contains polyclonal antibodies targeting the endotoxin LPS and other bacterial components (15). IMM-124E contains 80% proteins, out of which approximately 35% to 40% are immunoglobulins (15). IMM-124E is harvested from the colostrum of dairy cows that have been immunized against the outer antigens, mostly LPS, of the most common strains of ETEC. In Australia, Travelan is a listed medicine in the Australian Register for Therapeutic Goods (AUST L 106709) and is specifically indicated to reduce the risk of Travelers’ Diarrhea and reduce the symptoms of minor GI disorders (12).

TABLE 1.

Enterotoxigenic E. coli (ETEC) strains in inactivated ETEC vaccines (Immuron Ltd.)

Serotype	Strain no.	Source	Date of isolation
ETEC O6: H16	B2C	USA	Pre1971
ETEC O8: H19	C55 3/3c3	USA	mid1980s
ETEC O15: H4	PE 595	IMVS. Adelaide	Aust. Source
ETEC O25: H42	E11881A	USA	1986
ETEC O27: HR	C1067-77	USA	1985
ETEC O63: H-	PE 673	IMVS. Adelaide	Aust. Source
ETEC O78: H11	H10407	USA	Pre-1973
ETEC O114: H21	E20738/0	UK	1980
ETEC O115: H-	PE 724	IMVS. Adelaide	Aust. Source
ETEC O128: H21	EI 37–2	USA	1985
ETEC 148: H28	B7A	USA	Pre-1971
ETEC O153: H12	E8772/0	UK	Pre-1980
ETEC O159: H-	PE 768	IMVS. Adelaide	Aust. source

The hyperimmune bovine colostrum containing the Travelan LPS antibodies listed in Table 1 used in this study are referred to as “Travelan LPS antibody treatment”.

Mice used in the study

To assess for changes in mouse GI microbial populations following ingestion of the Travelan LPS antibody treatment/placebo, 20 C57BL/6 male mice aged 4 weeks were imported from the Animal Resource Centre in Western Australia to RMIT University, Bundoora and randomly allocated to a placebo or treatment group.

Feeding protocol

Four-week-old mice were habituated for 2 days and randomly allocated to a placebo or treatment group and placed in a group housing of 5 mice per cage. Ear notching was used to identify individual mice, so mice within the same treatment groups could be randomly moved between cages after each isolation period to control for the cage effect. Mice were weighed prior to commencing the feeding protocol (days 0, 1, and 2) and at the termination of the feeding protocol (day 8) (Fig. S1). Researchers were not blinded to treatment.

Mice were not food-deprived prior to the feeding experiment. Standard chow (Specialty Feeds Irradiated Rat and Mouse diet cubes, Specialty Feeds, Glen Forrest, and Western Australia) and water were available ad libitum during the 2 days prior to the implementation of five consecutive days of Travelan LPS antibody treatment or placebo. In the group-housed cages, the standard chow was removed from 9 a.m. to 4 p.m. for the 5 days of the feeding protocol. During the food restriction period, mice were placed in individual open-topped cages for 1 h twice a day (from 9 to 10 a.m. and 3–4 p.m.) and were given access to the LPS antibody treatment or placebo solutions. Mice within the placebo group were given access to 2 mL of freshly made 10% weight/volume Promilk solution, and mice in the treatment group were given access to 2 mL of freshly made 10% weight/volume Travelan LPS antibody solution. The water-soluble powders were reconstituted before the start of each feeding session by dissolving 3 g of powder in 30 mL dH₂0. The mixture was then stirred slowly for up to 5 min using a magnetic stirrer to avoid damaging the proteins. A 30 mM diameter plastic petri dish containing either placebo or Travelan LPS antibody treatment was attached to the floor of the cage with double-sided tape to allow mice free access to the solution for the 1 h treatment period. After each treatment period, mice were randomly returned to group housing (5 mice per cage in a total of 4 cages), the remaining treatment/placebo solution was weighed, and the amount consumed by each mouse per day was calculated.

Sample collection

Fresh fecal samples were collected using sterile forceps, placed in individual sterile 1.5 mL Eppendorf tubes, snap-frozen in liquid nitrogen and stored at −80°C. Fecal samples were collected at the end of each isolation period twice daily for 2 days prior to commencing the feeding protocol, and then twice daily for the 5 days feeding protocol. Individual open-topped cages were cleaned using 70% ethanol at the end of each isolation period.

At the end of the protocol, all 20 mice were culled by cervical dislocation (Ethics approval #1810, RMIT University) and final body weight was recorded. Gut tissue was dissected and anatomical measurements (cecal weight, length of small intestine, and colon) were recorded. Cecal microbial content was transferred directly into sterile tubes, colonic mucosal scrapings were collected aseptically from each mouse and snap-frozen in liquid nitrogen for microbial analysis.

Microbial DNA extraction

Microbial genomic DNA was extracted from approximately 0.25 g of fecal, cecal, and colonic samples using the QIAGEN DNeasy PowerSoil DNA isolation kit (QIAGEN, Hilden, Germany) as per manufacturer’s instructions. DNA concentrations were measured using a Qubit dsDNA HS Assay kit and Qubit 3.0 Fluorometer (ThermoFisher Scientific; Invitrogen, MA, USA). All samples were stored at −30°C until required.

Quantitative PCR

Quantitative PCR (qPCR) was used to quantify the total bacterial DNA copy number as an indicator of abundance. The primer pair 1114 f-1275r, which targets the bacterial 16S rRNA gene (23) was used to detect bacterial communities. The qPCRs were run on a CFX Connect Real-Time PCR Detection System (Bio-Rad). Each 5 µL reaction solution contained 0.25 pM of each forward and reverse primer, 2.5 µL of SensiFAST SYBR & Fluorescein mix (Bioline), sterile DNA-free water and 5 ng of sample DNA. Thermocycling conditions were 20 s at 95°C, followed by 40 cycles of 95°C for 3 s, and 61.5°C for 30 s (24). Reactions were followed by a melting curve increasing 1°C every 30 s, from 60°C to 99°C. Bacterial copy number was quantified by using 1114 f-1275r primers to amplify the 16 s rRNA gene from E. coli (DH5α). Standard curves were generated using triplicate 10-fold dilutions of the E. coli purified amplicon.

16s rRNA metabarcoding and bioinformatics

The V3–V4 hypervariable regions of the bacterial 16S rRNA gene (forward primer 5′ TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG 3′; reverse primer 5′ GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC 3’) were amplified on an Illumina MiSeq platform (2 × 300 bp) following the Illumina 16S metagenomic sequencing library preparation protocol (25, 26). Raw, demultiplexed, FASTQ files were re-barcoded, joined, and quality filtered following the UPARSE operational taxonomic unit clustering pipeline (27) (USEARCH; http://drive5.com/uparse/). Joined paired-end reads were quality filtered by discarding reads with total expected errors > 1.0. Reads that could not be assembled were discarded. Sequences were denoised and the -unoise3 command was used to generate zero-radius OTUs (ZOTU) (28, 29). Taxonomic assignments were performed using the USEARCH SINTAX algorithm (30). Reference databases were created using the RDP_trainset_15 data sets available from UTAX (https://www.drive5.com/usearch/manual8.1/utax_downloads.html). The minimum percentage identity required for a ZOTU to be considered a database match a hit was 90%. A phylogenetic tree was generated using MUSCLE (31). ZOTUs identified as chloroplasts and mitochondrial DNA were removed from the data. For clarity, ZOTUs clustered to higher taxonomic levels such as genus, will be referred to as OTUs.

Statistical analysis

Anatomical analysis

Statistical analyses were performed using GraphPad Prism v9.0.1 for Windows (GraphPad Software, San Diego, California, USA). Cecal weight, small intestinal length, and colon length taken at the time of culling were adjusted for individual body weight. Anatomical measurements were analyzed using two-tailed, Student’s t-test to assess for differences between Travelan LPS antibody-treated and placebo-treated mice, and P < 0.05 was taken to indicate statistical significance. Data are presented in a box and whisker plot, displaying the mean, median, interquartile range, and range of data.

Microbial analysis

Statistical analyses were performed in the R statistical computing software environment (version 4.0.5) (32). We conducted an alpha-diversity analysis to assess for differences in microbial communities across treatment groups. To determine whether there were differences in community structure between the placebo and Travelan LPS antibody treatment microbial communities, we analyzed beta diversity. Alpha- and beta-diversity analyses were performed on OTU-matrices rarefied to a depth of 10,000 reads using the “phyloseq” (33) and “vegan” (34) packages.

Normality and variance homogeneity of the data were tested using the “shapiro.test” and “bartlett.test” functions. Alpha diversity metrics were assessed using t-tests or, where appropriate, analysis of variance (ANOVA) with posthoc Tukey honest significant difference (HSD) tests. Nonmetric multidimensional scaling (NMDS) ordinations were used to visualize community beta diversity. As this study aimed to identify biomarkers that were responding to LPS treatment, the dissimilarity matrices were created using the weighted and unweighted UniFrac distance metric (35, 36). Significance testing of beta diversity was carried out using permutational MANOVAs (PERMANOVA) tests of factors affecting community structure (37).

Differential abundance testing of OTUs was performed using the DESeq2 extension available within the “phyloseq” package (33, 38). The differential abundance tests were performed on un-normalized community data as recommended by McMurdie and Holmes (39), for each taxonomic rank to identify which individual OTUs were changing in response to Travelan LPS antibody treatment. P values were corrected for multiple testing, using the Benjamini-Hochberg false discovery rate controlling procedure (40).

RESULTS

To characterize the effects of colostrum-containing LPS antibodies on the microbiome of unchallenged animals, mice were fed a placebo (milk powder) solution or a solution containing Travelan LPS antibody treatment, and fecal microbes were collected to enable microbial population comparisons. Specifically, we aimed to test the onset and magnitude of the effect of hyperimmune bovine colostrum containing ETEC LPS antibodies on the mouse gut microbiome by assessing microbial richness and diversity in fecal, colon mucosal, and cecal samples using deep sequencing methods.

Effects of LPS antibody-containing colostrum on GI anatomy

Prior to assessing for changes in mouse GI microbial populations following LPS antibody colostrum supplementation, we first sought to determine if the treated mice exhibited any gross anatomical differences compared with placebo-fed mice. At the end of the feeding protocol, all 20 mice were culled and anatomical measures (cecal weight, small intestinal length, and colon length) were taken at the time of culling. These measures were then adjusted for individual body weight.

To determine if mice fed Travelan LPS antibody treatment have altered gut structure compared to placebo-fed mice, body weight (Fig. S3), and gross anatomical measures of the GI tract were recorded from freshly culled mice. All 20 mice consumed the milk or treatment solution each day of the experiment. No significant difference in the consumption of placebo/ Travelan LPS antibody solution between treatment groups for the duration of the protocol was observed (Fig. S2). All mice in both placebo and LPS antibody colostrum-supplemented groups gained weight during and after the feeding protocol (Fig. S3 and S4). There were no significant differences in cecal weight normalized to body weight between placebo-fed mice and Travelan LPS antibody-treated mice (Fig. 1; P = 0.140). No significant differences were identified in small intestine length normalized to body weight between placebo-fed mice and Travelan LPS antibody-treated mice (Fig. 1; P = 0.452). Raw data showed no difference in colon, small intestinal length, or cecal weight (Fig. S5). A trend towards increased colon length in LPS antibody-treated mice compared to placebo-fed mice was observed (Fig. 1, P = 0.056).

Fig 1.

Gastrointestinal tract anatomical measures normalized to body weight. Left, mouse cecal weight (P = 0.140); center, small intestinal (SI) length (P = 0.452); right, colon length (P = 0.056); as a proportion of total body weight. Placebo: Promilk 85, Tatura Milk Industries Ltd, Australia; LPS-Ab: Travelan LPS antibody colostrum solution. Boxplots represent median, interquartile range and range of the data; cross inside the boxplot represents the mean.

Total bacterial quantification

To investigate the impact of Travelan LPS antibody treatment on bacterial numbers, bacterial diversity and species richness, microbial samples from different regions of the GI tract were collected from mice following completion of the feeding protocol.

We used quantitative PCR (qPCR) and next-generation sequencing to investigate if Travelan LPS antibody treatment induced changes to the diversity and species richness of bacterial communities from fecal, colonic mucosal, and cecal samples. Our qPCR data showed that Travelan LPS antibody treatment had no effect on bacterial cell numbers in any of the GI community samples. Although, we initially saw an increase in bacterial cell number in fecal samples for both groups of mice, this number had begun to return to the initial values by the end of the feeding protocol, resulting in no significant difference in total copy number between the treatment groups, or when compared to before the feeding protocol commenced (Fig. 2).

Fig 2.

qPCR estimates of total bacterial cell number based on 16S copy number for feces, colon mucosa, and cecum samples. Fecal samples (panel A) from days 3 (pretreatment), 5 (early treatment), and 8 (final treatment) of the experiment. Letters indicate statistical difference in copy number per microgram of DNA between sampling times (Days). Bacterial numbers collected from days with different letters are significantly different (treatment effect P > 0.05, time effect (Collection Day) P < 0.05, two-way ANOVA, and panel A). The total number of bacterial cells in mucosal (panel B) and cecal samples (panel C) at day 8 (end of experiment) were similar. Placebo: Promilk 85, Tatura Milk Industries Ltd, Australia); LPS-Ab: Travelan LPS antibody treatment. Boxplots represent median and interquartile range; whiskers represent 1.5 x interquartile range (Tukey method).

Community richness and diversity

Across fecal communities and colonic mucosal content collected post-mortem, there was no significant effect of time or Travelan LPS antibody treatment on bacterial diversity (Shannon-Wiener diversity), OTU (Operational Taxonomic Units) richness, or evenness (Fig. 3). However, community evenness was reduced in the cecal communities of mice supplemented with the Travelan LPS antibody treatment when compared to controls (P = 0.009, Fig. 3).

Fig 3.

Bacterial community diversity metrics; Shannon diversity (top), Chao1 richness (middle), and Simpson’s evenness (bottom). Fecal samples were collected pretreatment (Day 3) and after treatment (Day 5 and 8). Day 3 represents the first day of feeding the mice with the Travelan LPS antibody treatment (LPS-Ab)/placebo solution (n = 10 placebo, n = 10 LPS-Ab). Colon samples were collected post-mortem (Day 8) and include primarily mucosal content, with some fecal content n = 10 placebo, n = 10 LPS-Ab. Cecal samples were collected post-mortem (Day 8) n = 10 placebo, n = 10 LPS-Ab. Placebo: Promilk 85, Tatura Milk Industries Ltd, Australia); LPS Ab: Travelan LPS antibody treatment. Boxplots represent median and interquartile range; whiskers represent 1.5 x interquartile range (Tukey method).

Travelan LPS antibody treatment significantly impacted the structure of fecal communities as detected by Two-way PERMANOVAs (pseudo-F = 2.3702, P = 0.031) using weighted UniFrac distances. There was no detectable effect of sampling time on microbiota. LPS antibody colostrum treatment also resulted in significant grouping of bacterial communities for both colonic mucosal (pseudo-F = 2.2432, P = 0.041) and cecal (pseudo-F = 3.4661, P = 0.005) communities (Fig. 4; Table S1). Changes in response to treatment were also observed in unweighted UniFrac ordinations (Fig. S6; Table S1).

Fig 4.

Non-metric multidimensional scaling (NMDS) ordination based on weighted UniFrac distances showing relationship between bacterial community structure from fecal samples from Days (D) 3, 5, and 8 of the experiment (left, pseudo-F = 2.3702, P = 0.031) and colonic mucosal samples (center, pseudo-F = 2.2432, P = 0.041) and cecum samples (right, pseudo-F = 3.4661, P = 0.005) from D8 (end of experiment). Circles = placebo (Promilk 85, Tatura Milk Industries Ltd, Australia) ; Triangles = Travelan LPS antibody treatment.

In summary, these data suggest that Travelan LPS antibody treatment had minimal to no impact on bacterial cell number, bacterial species richness, or species abundance distribution (evenness) in fecal and colon mucosal samples. In contrast, in cecal samples, there was a reduction in community evenness found in the Travelan LPS antibody-treated samples compared to placebo-treated samples, indicative of selection pressure acting on these communities. Most likely, decreases in rare taxa produced a relative increase in dominant taxa, lowering community evenness. Cecal samples from LPS antibody-treated mice exhibited a trend for an overall reduction in species diversity compared to controls. At the community level, small but significant shifts in community structure were observed for fecal, mucosal, and cecal samples following treatment.

Travelan LPS antibody treatment results in differentially abundant bacterial OTUs

Given that Travelan LPS antibody treatment impacted bacterial community structure in fecal, mucosal, and cecal samples, differential abundance testing of OTUs was performed for each taxonomic rank to identify which individual OTUs were changing in response to the treatment. Across fecal samples, the genera Akkermansia and Enterorhabdus were enriched in Travelan LPS antibody-treated communities and microbes within the genus Clostridium (Clade III) were reduced (Table 2). Akkermansia displayed the largest significant change in abundance of all genera irrespective of sample type. For mucosal communities, no individual OTUs or genera differed significantly in their abundance, however, there was a trend towards an increase in Olsenella and Desulfovibrio, and a decrease in Clostridium (Clade XVIII) (Table 3). In cecal content communities, Travelan LPS antibody treatment decreased the abundance of eight bacterial genera, in particular; Porphyromonas, a producer of short-chain fatty acids (41), and Parabacteroides which contributes to bile acid metabolism and plays a role in reinforcing the enteric epithelium (42, 43) (Table 4). At higher taxonomic levels (Family and Order), the Desulfovibrionales have significantly enriched Travelan LPS antibody-treated mucosal communities.

TABLE 2.

Genus-level OTUs with altered relative abundances for fecal communities ^a

Relative abundance	Log2FC	Effect of LPS Ab	ADJ P-value	Phylum	Order	Genus
0.27	6.09	Increase	7.86E-06	Verrucomicrobia	Verrucomicrobiales	Akkermansia
0.77	1.25	Decrease	2.56E-02	Firmicutes	Clostridiales	Clostridium_III
0.07	1.03	Increase	7.44E-03	Actinobacteria	Coriobacteriales	Enterorhabdus

^a The table indicates the relative abundance of each significantly different taxa across the entire data set, as well as the change in abundance between the treatment and control groups, presented as the log2 fold change (LOG2FC). The impact of Travelan LPS antibody treatment on the abundance of each taxon is also indicated as either decreasing or increasing in response to the treatment.

TABLE 3.

No significant alteration to the relative abundance of genus-level OTUs in colon mucosal communities ^a

ReLative abundance	Log2FC	Effect of LPS Ab	ADJ P-value	Phylum	Order	Genus
0.20	3.73	Decrease	0.057	Firmicutes	Erysipelotrichales	Clostridium XVIII
0.46	1.62	Increase	0.075	Proteobacteria	Desulfovibrionales	Desulfovibrio
0.21	4.23	Increase	0.057	Actinobacteria	Coriobacteriales	Olsenella

^a The table indicates the relative abundance of borderline significantly different taxa across the entire data set, as well as the change in abundance between the treatment and control groups, presented as the log2 fold change (LOG2FC). The impact of Travelan LPS antibody treatment on the abundance of each taxon is also indicated as either decreasing or increasing in response to the treatment.

TABLE 4.

Altered relative abundance of genus-level OTUs for cecal communities in response to travelan LPS antibody treatment ^a

Relative abundance	Log2FC	Effect of LPS Ab	ADJ P-value	Phylum	Order	Genus
0.35	3.47	Decrease	0.014	Firmicutes	Erysipelotrichales	Allobaculum
17.55	1.12	Increase	0.061	Firmicutes	Clostridiales	Clostridium XIVa
3.55	1.25	Decrease	0.014	Bacteroidetes	Bacteroidales	Parabacteroides
0.52	1.93	Decrease	0.005	Bacteroidetes	Sphingobacteriales	Parapedobacter
0.03	2.43	Decrease	0.010	Bacteroidetes	Sphingobacteriales	Pedobacter
1.10	1.64	Decrease	0.001	Bacteroidetes	Bacteroidales	Porphyromonas
0.03	2.54	Decrease	0.014	Bacteroidetes	Flavobacteriales	Robiginitalea
0.26	2.34	Decrease	0.004	Bacteroidetes	Bacteroidales	Saccharicrinis
0.02	2.53	Decrease	0.040	Bacteroidetes	Sphingobacteriales	Sphingobacterium

^a The table indicates the relative abundance of each significantly different taxa across the entire data set, as well as the change in abundance between the treatment and control groups, presented as the log2 fold change (LOG2FC). The impact of Travelan LPS antibody treatment on the abundance of each taxon is also indicated as either decreasing or increasing in response to the treatment.

There was no significant difference in the relative abundances of the OTUs in the colon mucosal samples of the control and mice treated with Travelan LPS antibody treatment.

DISCUSSION

The human gut is a major reservoir of LPS resulting from the shedding of the outer layer of gram-negative bacteria. Elevated LPS levels promote intestinal inflammation through the release of pro-inflammatory cytokines (44) in conditions such as necrotizing enterocolitis (45), inflammatory bowel disease (46), obesity (47), and nonalcoholic fatty liver disease (48). The immunological activity of IgGs in hyperimmune colostrum raised against a bacterial LPS extract (Imm124-E, Travelan, Immuron Ltd., Melbourne, Australia) is of increasing relevance to human health due to its ability to directly target specific pathogens (12, 15), protect against ETEC-induced diarrhea (12), reduce inflammation by preventing translocation of bacterial antigens across the epithelium (12, 49), and inhibit the induction of proinflammatory cytokines (16, 49, 50) in human trials and murine models.

Here, we show that a treatment rich in LPS antibodies (15) can engineer GI bacterial communities, without dramatically disrupting the balance and underlying ecology of these GI communities. The effects of colostrum antibodies targeting LPS on the gut microbiome are poorly studied. Therefore, we aimed to assess the onset and magnitude of changes in the richness and diversity of gut microbial populations in unchallenged mice administered a therapeutic targeting LPS. Overall, the treatment had minimal impact upon the total number of bacterial cells, bacterial species, community diversity, and evenness in a healthy preclinical model.

We first aimed to investigate for changes in GI anatomical parameters following completion of feeding mice milk powder solution (placebo), or a solution containing Travelan LPS antibody treatment. We demonstrate that administration of antibodies did not affect the length of small intestine or cecal weight in mice. Interestingly, mice fed LPS antibodies showed a strong trend towards longer colons compared to control mice. These findings suggest no adverse change in GI anatomy due to dietary supplementation with Travelan LPS antibodies.

We showed that although bacterial cell number, bacterial species richness, or species abundance distribution (evenness) in fecal and mucosal samples remained unchanged following Travelan LPS antibody colostrum supplementation, bacterial evenness in cecal samples was reduced compared to samples from placebo-treated mice. In addition, treatment with colostrum containing Travelan LPS antibodies caused a small but significant shift in microbial community structure in fecal, colon mucosal, and cecal samples indicating that microbial communities were altered along the GI tract, which may contribute to subtly altered GI function.

To understand which microbes were potentially driving the observed changes in community structure, we investigated changes at the level of bacterial OTUs. Using this approach, we revealed shifts in numerous low-abundance bacterial genera, however, many of these are poorly characterized in the GI tract. Minor changes to rare and low-abundance species may have a larger impact than anticipated if unique functions provided by these species are disrupted, especially if these species are important members of the ecological core (51). Further exploration of the roles provided by rare species in the GI tract, as well as the relationship between these species is required.

The largest change to the fecal bacterial community in response to Travelan LPS antibody treatment was an increase in abundance of the genus Akkermansia. Akkermansia is involved in maintaining intestinal barrier integrity, reducing inflammation, regulating host metabolism and plays a role in the immune tolerance of commensal organisms (52 – 55). It is widely associated with healthy mucus-associated microbial communities and is a prolific re-colonizer of the GI tract after disturbances such as infection (52). Based on the findings of the current study, we speculate that Travelan LPS antibody treatment creates ecological niches for Akkermansia to proliferate into by knocking down/removing other bacterial species whose abundances were below the detection threshold of our current statistical approaches.

In Travelan LPS antibody-treated colonic mucosal communities, an increase of Desulfovibrionales was observed at higher taxonomic levels (i.e., Family and Order). Elevated levels of Desulfovibrio, a sulfate-reducing bacteria, are commonly associated with a penetrable mucus phenotype in animals exhibiting gut inflammation (56, 57). While the increase in Desulfovibrionales may indicate the presence of an inflammatory response (58, 59), the change may also be attributable to the important metabolic role Desulfovibrio plays in the GI tract. The utilization of excess hydrogen through the processes of sulfate reduction, methanogenesis, and acetogenesis by hydrogen cross-feeders assists in the maintenance of gut homeostasis (60). Desulfovibrio is the most dominant group of sulfate-reducing bacteria in the colon (61, 62); utilizing hydrogen to convert sulfate to sulfide compounds while oxidizing lactate to acetate (63), a beneficial short-chain fatty acid. Interestingly, hydrogen and the products of hydrogen-feeders are associated with both health-promoting effects such as mucus layer integrity (64, 65), and detrimental health outcomes observed in Parkinson’s disease and inflammatory bowel disease (66 – 69). This apparent contradiction may reflect differences in the microbial species composition involved in hydrogen cycling within individuals. In addition, the sulfate utilized by sulfate-reducing bacteria like Desulfovibrio may be derived from dietary sources of sulfur or released from the breakdown of endogenous mucins by other microorganisms, such as Akkermansia (70, 71). Therefore, an increase in Desulfovibrionales without evidence of pathology may merely indicate an increase in substrate availability due to an increased abundance of mucin degraders like Akkermansia.

Although several changes were observed in cecal communities due to the administration of Travelan LPS colostrum, only community members of relatively rare abundance were impacted, and the abundance changes were minor. We observed a reduction in cecal microbial community evenness indicating that Travelan LPS antibody-treated cecal communities have more dominant OTUs. This may be due to the trend for an increase in Clostridium XIVa abundance in Travelan LPS antibody-treated cecal samples, which was marginally, but not statistically significant. These community diversity metrics incorporate both OTU richness and evenness. Our finding that the cecal community diversity was unaltered despite a shift in evenness, suggests that community evenness changes induced by treatment with Travelan LPS antibodies are minimal, although statistically significant.

We acknowledge that a limitation of this study is the immunomodulatory components present in colostrum, such as growth factors and cytokines, that were not present in the control and may have contributed to the observed changes in community composition. However, collectively, these data suggest that LPS colostrum treatment administered in the absence of infection does not adversely affect the GI microbial community structure and has very little cross-reactivity towards commensal bacteria. The ability to target specific bacterial species, without large off-target effects, makes this a promising tool in the selective modification of GI microbial composition in diseases with GI co-morbidities. Further investigation is needed to determine whether these subtle shifts in abundance translate to a change in the functional attributes of these microbial communities.

Conclusions

This is the first study to assess the impact of Travelan LPS antibodies on anatomical and microbial changes in the wild-type nonimmune challenged mouse GI tract. There was no difference in any of the anatomical measures obtained between groups, however, we did identify small but significant differences in rare bacterial species of mouse GI microbial populations following treatment with Travelan LPS antibodies. Specifically, increases in the abundance of the gut bacterial genera Akkermansia were identified in fecal samples as well as reductions in the abundance of eight relatively rare OTUs in the cecal microbial community. The increase in Akkermansia may be a beneficial off-target outcome caused by the reduction in rare bacterial species induced by Travelan LPS antibody treatment. Our results indicate that administering Travelan LPS antibody treatment to a healthy preclinical model can modulate GI bacterial communities without dramatically disrupting the balance or underlying ecology of GI communities. Further studies are required to examine the potential of Travelan LPS antibody treatment to engineer the microbiome to alleviate gut dysbiosis in preclinical models and humans.

DATA AVAILABILITY

The datasets presented in this study can be found in online repositories. The names of the repositories are as follows: NCBI BioProject (accession number PRJNA785752, RMIT University figshare repository (https://doi.org/10.25439/rmt.17141858.v1)

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/iai.00097-23.

Supplementary Material. iai.00097-23-s0001.pdf.

6 supplemental figures and 1 supplemental table combined in PDF format.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.