Protective role of commensal bacteria in Sjögren Syndrome

Abstract

CD25 knock-out (CD25KO) mice spontaneously develop Sjögren Syndrome (SS)-like inflammation. We investigated the role of commensal bacteria by comparing CD25KO mice housed in conventional or germ-free conditions. Germ-free CD25KO mice have greater corneal barrier dysfunction, lower goblet cell density, increased total lymphocytic infiltration score, increased expression of IFN-γ, IL-12 and higher a frequency of CD4+IFN-γ+ cells than conventional mice. CD4+ T cells isolated from female germ-free CD25KO mice adoptively transferred to naive immunodeficient RAG1KO recipients caused more severe Sjögren-like disease than CD4+ T cells transferred from conventional CD25KO mice. Fecal transplant in germ-free CD25KO mice reversed the spontaneous dry eye phenotype and decreased the generation of pathogenic CD4+IFN-γ+ cells. Our studies indicate that lack of commensal bacteria accelerates the onset and severity of dacryoadenitis and generates autoreactive CD4+T cells with greater pathogenicity in the CD25KO model, suggesting that the commensal bacteria or their metabolites products have immunoregulatory properties that protect exocrine glands in the CD25KO SS model.

Keywords: CD25KO; Dacryoadenitis; Dry eye; Fecal transplant; Germ-free; Microbiome; Sjögren Syndrome.

Conflict of interest statement

Conflict of interest

Baylor College of Medicine has filed a provisional patent. No other conflict of interest that could be perceived to bias the work.

Copyright © 2018 Elsevier Ltd. All rights reserved.

Figures

Figure 1. Germ-Free Environment Worsens Sjögren Syndrome in the CD25KO mice

Conventionally (CON) housed CD25KO mice were compared to germ-free (GF) CD25KO mice at four and eight weeks of age. Both sexes were used. (A–B) Representative pictures of the corneal permeability (A) and accumulative data (B) Corneas were stained with fluorescent Oregon-Green dextran (OGD) dye and OGD intensity score was calculated in the 2-mm central cornea by two masked investigators. Bar graphs show means ± SEM of two independent experiments with four to ten animals per experiment (final n = nine to twenty animals/group). Parametric t-test were used to make comparisons between groups (p value noted by asterisks). (C) Number of PAS+ conjunctival goblet cells counted in paraffin-embedded sections of 8-week old conventional and germ-free CD25KO mice. Bar graphs show means ± SEM of two independent experiments with three animals per group (final n = six right eyes for each group). Dotted line indicates goblet cell density in wild-type animals. Parametric t- tests were used to make comparisons between groups. (D) Total lacrimal gland score measured in H&E stained sections. Bar graphs show means ± SEM of three independent experiments with three-six animals per experiment (n = nine to eighteen right lacrimal glands). Nonparametric Mann–Whitney U tests were used to make comparisons of inflammation scores. (E) Representative pictures of haematoxylin and eosin (H&E)-stained sections of lacrimal gland. White quadrants insets are high magnification of black lined squares. 10X objective. Smaller quadrants are focused on acini appearance. (F) Means ± SD of gene expression analysis in germ-free and conventional lacrimal gland lysates. Bar graphs show means ± SD of two independent experiments with a final sample size of five to ten samples per group/age. Two-way Anova with Sidak’s post hoc test was used for comparison. (G) Flow cytometric analysis of lacrimal gland stained for CD4, CD8 and B220 at 8 weeks of age. Bar graphs show means ± SD of two independent experiments, with a final sample size of six samples per group/age. Parametric t-tests were used to make comparisons between groups. (H) Flow cytometric analysis of intracellular staining of lacrimal glands and cervical lymph nodes of conventional and germ-free mice at 8 weeks of age. Right and left extraorbital lacrimal glands from one mouse were excised and pooled into a single tube, yielding a final sample of nine to sixteen individual lacrimal gland samples or nodes divided into four independent experiments with three to-four samples per experiment. Bar graphs show means ± SEM. Parametric t-tests were used to make comparisons between groups. *P

Figure 2. Adoptive Transfer Recipients of Germ-free CD25KO CD4+T cells Develop SS-like Disease

CD4+ T cells were isolated from spleens and cervical lymph nodes (CLN) from either germfree (GF) CD25KO or conventional (CON) CD25KO mice and adoptively transferred (AT) i.p. into RAG1KO recipients (AT→ RAG). Ocular and lacrimal gland phenotype in RAG1KO recipients were investigated 5 weeks post-transfer. (A) Corneal permeability measured as an uptake of fluorescent Oregon-Green dextran (OGD) dye. Bar graphs show means ± SEM of two independent experiments with eight to ten animals per experiment (final n = eighteen to twenty animals). Parametric t- tests were used to make comparisons between groups. (B). Representative pictures used to generate the bar graphs in A. (C) Number of PAS+ conjunctival goblet cells counted in paraffinembedded sections expressed as number per millimeter. Bar graphs show means ± SEM of two independent experiments with two to three animals per group, yielding a final sample of five right eyes for each group. Parametric t- tests were used to make comparisons between groups. (D) Inflammation scores of lacrimal gland pathology of donor and AT recipients. Bar graphs show average of two independent experiments (final n = thirteen right lacrimal glands animals). Nonparametric Mann–Whitney U test was used to make comparisons of inflammation scores. (E) Representative pictures of haematoxylin and eosin (H&E)-stained sections of lacrimal gland in adoptive transfer RAG1KO recipients; 20X objective, scale bar = 50 μm. (F) Flow cytometric analysis of intracellular staining of lacrimal gland and cervical lymph nodes of adoptive transfer recipients (n= eight to fourteen individual right lacrimal glands). Bar graphs show means ± SEM of three independent experiments. Parametric t-tests were used to make comparisons between groups. (G) Gene expression analysis in lacrimal gland lysates of germ-free and conventional CD25KO adoptive transfer recipients. (AT→ RAG). Bar graphs show means ± SD of two independent experiments (n = eight lacrimal glands/group). *P

Figure 3. Fecal Microbiota Transplant in Germ-free CD25KO Ameliorates the Autoimmune Phenotype

Germ-free (GF) CD25KO were removed from isolators at four weeks of age and received a fecal transplant (FT) of fecal microbiota and were kept in normal vivarium until they reached 8 weeks of age. CD4+ T cells were isolated from spleens and cervical lymph nodes (CLN) and adoptively transferred into RAG1KO recipients (AT→ RAG). Ocular and lacrimal gland phenotype was investigated in both donor mice and RAG1KO recipients. (A–D) Donor mice phenotype after fecal transplant. (A) Corneal permeability measured as an uptake of fluorescent Oregon-Green dextran (OGD) dye. Bar graphs show means ± SEM of three independent experiments (final n = eighteen to twenty-one animals). Parametric t- tests were used to make comparisons between groups. (B) Number of PAS+ conjunctival goblet cells counted in paraffin-embedded sections. Bar graphs show means ± SEM of two independent experiments, final n = five right eyes for each group. Parametric t- tests were used to make comparisons between groups. (C) Inflammation scores of lacrimal gland pathology of donor and AT recipients. Bar graphs show average of two independent experiments (final n = ten animals). Nonparametric Mann–Whitney U statistical tests were used to make comparisons of inflammation scores. (D) Representative pictures of haematoxylin and eosin (H&E)-stained sections of lacrimal gland in donor mice. (E–J) Dry eye phenotype in RAG1KO recipients. (E) Corneal permeability measured as an uptake of fluorescent Oregon-Green dextran (OGD) dye. Bar graphs are means ± SEM of two independent experiments (final n = eight to ten animals). Parametric t- tests were used to make comparisons between groups. (F) Number of PAS+ conjunctival goblet cells counted in paraffin-embedded sections expressed as number per millimeter. Bar graphs show means ± SEM of two independent experiments, final n = five right eyes for each group. Parametric t- tests were used to make comparisons between groups. (G) Inflammation scores of lacrimal gland pathology in AT recipients. Bar graphs show average of two independent experiments (final n = ten animals). Nonparametric Mann–Whitney U statistical tests were used to make comparisons of inflammation scores. (H) Representative pictures of haematoxylin and eosin (H&E)-stained sections of lacrimal gland in adoptive transfer RAG1KO recipients; showing two mice per group. Black quadrants insets are high magnification of smaller black demarcated area. (I) Flow cytometric analysis of cervical lymph nodes and lacrimal gland of adoptive transfer RAG1KO recipients. Bar graphs show means ± SEM of two independent experiments (final n = seven to eleven animals/group). Parametric t-tests were used to make comparisons between groups.(J) Gene expression analysis in lacrimal gland lysates of adoptive transfer recipients. Bar graphs show means ± SD of two independent experiments, final n = nine to ten mice/group. Parametric t- tests were used to make comparisons between groups. *P

Figure 4. Antibiotic Treatment Worsens Dacryoadenitis in CD25KO mice

Wild-type (WT) and CD25KO mice were subjected to a cocktail of oral antibiotics (ABX) for 4 weeks starting at 4 weeks of age and compared to CD25KO at 8 weeks of age that drank normal water (n = five/group). All groups were born and raised in conventional specific-free pathogen vivarium. (A) Representative pictures of haematoxylin and eosin (H&E)-stained sections of lacrimal glands. Large inset squares are high magnification of small insets, 10X objective, scale bar = 100 μm. (B) Inflammation scores of lacrimal gland pathology. Bar graphs show means ± SEM of two independent experiments, final n = five animals. Nonparametric Mann–Whitney U statistical tests were used to make comparisons.(C) Relative fold of expression of IFN-γ and IL-12 in lacrimal glands. Bar graphs show means ± SEM of five samples per group. Parametric t-tests were used to make comparisons between groups. ****P

Figure 5. IL-12 Neutralization in Germ-free CD25KO mice Improves Autoimmunity

A-Single cell suspensions were prepared from conventional and germ-free lacrimal glands (LG), cervical lymph nodes (CLN) or conjunctiva and stained with live/dead dye and CD45, IL-12, CD11c, and MHC II antibodies. Single cells, alive CD45+IL-12+ cells were gated and CD11c and MHC II were plotted. B–D-Germ-free CD25KO mice, aged four weeks of age, received either anti-IL-12 (α-IL-12) or isotype control (IC) antibody bi-weekly i.p. injections (200ug/injection) for a total of 4 weeks while kept in a germ-free isolator. Mice were used at eight weeks of age (n = six/group). (A) Flow cytometry analysis showing distribution of CD45+IL-12+ cells based on CD11c and MHC II expression; pie charts of means of four samples. Means and SD are shown below. Nonparametric Mann–Whitney U tests were used to make comparisons between germ-free and conventional mice, * P<0.05, ***P<0.001. (B) Representative pictures of haematoxylin and eosin (H&E)-stained sections of the lacrimal gland. 20X objective, scale bar = 50 μm. (C) Inflammation scores of lacrimal gland pathology. Bar graphs show means ± SEM of two independent experiments, final n = five animals. Nonparametric Mann–Whitney U tests were used to make comparisons. (D) Flow cytometric analysis of cervical lymph nodes and lacrimal gland. Bar graphs are means ± SD of two independent experiments, with a final sample size of six per group. Parametric t-tests were used to make comparisons between groups. LG = lacrimal gland; CLN = cervical lymph nodes; NS = non-significant