Mus musculus deficient for secretory antibodies show delayed growth with an altered urinary metabolome

Abstract

Background: The polymeric immunoglobulin receptor (pIgR) maintains the integrity of epithelial barriers by transporting polymeric antibodies and antigens through the epithelial mucosa into the lumen. In this study, we examined the role of pIgR in maintaining gut barrier integrity, which is important for the normal development in mice.

Methods: Cohorts of pIgR^-/- mice and their wildtype controls were housed under Specific Pathogen Free (SPF) conditions and monitored for weight gain as an indicator of development over time. The general physiology of the gastrointestinal tract was analysed using immunohistochemistry in young (8-12 weeks of age) and aged mice (up to 18 months of age), and the observed immunopathology in pIgR^-/- mice was further characterised using flow cytometry. Urinary metabolites were analysed using gas chromatography-mass spectrometry (GC-MS), which revealed changes in metabolites that correlated with age-related increase in gut permeability in pIgR^-/- mice.

Results: We observed that pIgR^-/- mice exhibited delayed growth, and this phenomenon is associated with low-grade gut inflammation that increased with ageing. The gross intraepithelial lymphocytic (IEL) infiltration characteristic of pIgR^-/- mice was redefined as CD8α⁺αβ⁺ T cells, the majority of which expressed high levels of CD103 and CD69 consistent with tissue resident memory T cells (T_RM). Comparison of the urinary metabolome between pIgR^-/- and wild-type mice revealed key changes in urinary biomarkers fucose, glycine and Vitamin B5, suggestive of altered mucosal permeability. A significant increase in gut permeability was confirmed by analysing the site-specific uptake of sugar probes in different parts of the intestine.

Conclusion: Our data show that loss of the secretory antibody system in mice results in enhanced accumulation of inflammatory IELs in the gut, which likely reflects ongoing inflammation in reaction to gut microbiota or food antigens, leading to delayed growth in pIgR^-/- mice. We demonstrate that this leads to the presence of a unique urinary metabolome profile, which may provide a biomarker for altered gut permeability.

Keywords: Gut; Inflammation; Permeability; SIgA; Tissue resident memory T cells; Urinary biomarker.

Fig. 1

Age-dependent immunopathology in the GIT of pIgR^−/− mice. a Intestine sections of 2-month and 18-month old male B6 and pIgR^−/− mice were stained with H&E. Representative sections of duodenum, ileum and colon were imaged at 20× magnification. Arrows indicate pathology in duodenum of 18-month old pIgR^−/− mice. b Intestine sections were graded blindly by a pathologist. Intestines were described as normal or affected. Affected tissue exhibited a reduction in mucin, larger nuclei in crypt epithelium and seemingly increased number of lymphocytes. Affected samples are noted in the table. Statistical analysis was performed using a two-tailed Fisher’s exact test, ns indicates p-values > 0.05

Fig. 2

Flow cytometric analysis of small intestine intra-epithelial lymphocytes (IELs) reveals enrichment of αβ⁺ T cell subsets in older pIgR^−/− mice. IELs were collected from age- and sex-matched B6 (closed circle, 12-week n = 7, 34-week n = 14) and B6.pIgR^−/− (open circle, 12-week n = 8, 34-week n = 11) mice. a The total number of small intestine IELs (SI-IELs, CD45⁺), αβ⁺ T cells (TCRβ⁺) and γδ⁺ T cells (TCRγδ⁺) was increased in pIgR^−/− mice compared to B6 mice. Representative FACS profiles of SI-IEL αβ⁺ T cell are shown for b MAIT cells by MR1 tetramer staining with SI-IELs from Vα19i MAIT cell transgenic mice as positive control for gating (Kawachi et al. 2006), and c NKT cells by CD1d tetramer staining with splenocytes as positive control for gating. The number of d SI-IEL CD8α⁺αβ⁺ T cells (TCRβ⁺CD8α⁺CD4⁻) and SI-IEL CD4⁺αβ⁺ T cells (TCRβ⁺CD4⁺) was significantly increased in pIgR^−/− mice especially with age, and e representative FACS profiles for these subsets are shown. f Serum was collected from B6 and pIgR^−/− mice at indicated age and small intestine was homogenised in 2 ml RPMI. The concentration of IFN-γ was determined using mouse Th1/Th2/Th17 cytometric bead array (CBA). Dotted line indicates detection limit (DL). Data points indicate individual mice that were pooled from 3 to 4 independent experiments, mean ± SEM is shown. Unpaired t-test was used for statistical analyses and p-values are shown where < 0.05 is considered significant

Fig. 3

Effect of pIgR deletion on mouse body weight in B6 mice. Cohorts of age- and sex-matched male (closed circle, B6 n = 15; open circle, pIgR^−/− n = 10) and female (closed diamond, B6 n = 10; open diamond, pIgR^−/− n = 10) mice were weighed at indicated age. Shown are a body weight of individual mice at 4 weeks of age; and b mean ± SEM body weight over a 65-week time course. c Proportional weight gain is calculated as weight gained since 4 weeks of age as a percentage of 4-week body weight of the same mouse

Fig. 4

Scores plot analysis of metabolite profiles in urine of B6 and pIgR^−/− mice. GC-MS was performed on urine samples to determine the profile of metabolites excreted in the urine of a male and b female B6 and pIgR^−/− mice. To control for sample variation due to urine volume and concentration, the annotated urine metabolites were normalized by Mean Absolute Deviation (MAD). Data was analyzed by sPLS-DA and scores plot, with 95% confidence intervals of the sample space highlighted by elliptical boundaries

Fig. 5

Specific metabolites in the urine of B6 and pIgR^−/− mice. GC-MS was performed on samples to determine the profile of metabolites excreted in urine of male (closed circle, B6 n = 17; open circle, pIgR^−/− n = 19) and female (closed diamond, B6 n = 9; open diamond, pIgR^−/− n = 10) mice. Samples were urease treated, ethanol extracted and analysed by GC-MS with myo-inositol as internal control and individual sugars quantitated using a standard sugar mix to determine response factors. Normalized peak area values are shown for a fucose, b glycine and c pantothenic acid. Unpaired t-test was used for statistical analyses and p-values are shown, ns indicates p-value > 0.05

Fig. 6

Gastrointestinal permeability of older adult mice to sugar probes. Gastrointestinal permeability was measured as altered ratios of sugar probes excreted in the urine. a B6 mice were treated with 5 mg aspirin two hours prior to feeding with sugar probes (open star sign, untreated n = 14; closed star sign, treated n = 6). b 10-week old or c 22-week old female B6 (closed diamond) or pIgR^−/− (open diamond) mice were given an oral dose of sugar probes containing sucrose, lactulose, sucralose and mannitol. The concentration of sugar probes in urine was collected 24-h post feeding and analysed by GC-MS. Data are represented as the ratio of sugar probe concentrations within samples and the horizontal bar represents the mean of each group. Statistical analysis was performed using the Mann Whitney U-test, ns indicates p > 0.05. Data are pooled from two independently performed experiments