Nutritional Support from the Intestinal Microbiota Improves Hematopoietic Reconstitution after Bone Marrow Transplantation in Mice

Abstract

Bone marrow transplantation (BMT) offers curative potential for patients with high-risk hematologic malignancies, but the post-transplantation period is characterized by profound immunodeficiency. Recent studies indicate that the intestinal microbiota not only regulates mucosal immunity, but can also contribute to systemic immunity and hematopoiesis. Using antibiotic-mediated microbiota depletion in a syngeneic BMT mouse model, here we describe a role for the intestinal flora in hematopoietic recovery after BMT. Depletion of the intestinal microbiota resulted in impaired recovery of lymphocyte and neutrophil counts, while recovery of the hematopoietic stem and progenitor compartments and the erythroid lineage were largely unaffected. Depletion of the intestinal microbiota also reduced dietary energy uptake and visceral fat stores. Caloric supplementation through sucrose in the drinking water improved post-BMT hematopoietic recovery in mice with a depleted intestinal flora. Taken together, we show that the intestinal microbiota contribute to post-BMT hematopoietic reconstitution in mice through improved dietary energy uptake.

Keywords: antibiotics; bone marrow cell transplantation; hematopoiesis; hematopoietic recovery; immune reconstitution; intestinal flora; microbiota; nutrition.

Fig. 1. Depletion of the intestinal microbiota impairs immune reconstitution after bone marrow transplantation and sensitizes mice to sub-lethal irradiation

(A) Experimental procedure of BMT. PB – Peripheral blood analysis. (B) Quantification of bacterial 16S rRNA in fecal samples from untreated control (n = 10), ampicillin + enrofloxacin (AE)-treated (n = 10), and vancomycin + amikacin (VA)-treated mice (n = 5) 14 days after BMT. NTC = Non Template Control. (C) White blood cells (WBC), red blood cells (RBC), platelets (PLT), lymphocytes (LYMPH), neutrophils (NEUT), and monocytes (MONO), (D) Flow-cytometry analysis of B and T cells in peripheral blood after BMT and (E) Total bone marrow cellularity 28 days after BMT in control (n = 10), AE-treated (n = 8), and VA-treated mice (n = 10). (F) Experimental procedure of semi-lethal irradiation. (G) Survival and (H) Representative images of hematoxylin- and eosin-stained bone-marrow vertebrae sections from untreated mice (left panel) and AE-treated mice (middle panel) 21 days after 750cGy radiation and from an age-matched untreated unirradiated control mouse (right panel). Scale bar 100μm. (I) Experimental procedure of control/resistant fecal microbiota transfer (FMT) and subsequent BMT. (J) Quantification of bacterial 16S rRNA copies in fecal samples from mice at day 0. Control (C) FMT with or without AE-treatment and resistant (R) FMT with or without AE-treatment (n = 5 per group). (K) Total bone marrow cellularity 28 days after BMT and (L) WBC, RBC, PLT, LYMPH, NEUT, and MONO counts after BMT in mice given a control FMT without (n = 10) or with (n = 9) AE-treatment and mice given a resistant FMT without (n = 10) or with (n = 10) AE-treatment. Significance levels are comparison of AE-treated Ctrl-FMT and AE-treated Res-FMT. Shaded areas in (C) and (L) indicate normal ranges. * P < 0.05, ** P < 0.01, *** P < 0.001, n.s. – Not significant. Results represent at least two independent experiments. Data is presented as mean ± SEM. See also Figure S1.

Fig. 2. Abx-mediated depletion of the intestinal microbiota predominantly suppresses hematopoietic differentiation

Spider plot (left panel) and heat map (right panel) of number of long-term hematopoietic stem cells (LT-HSC, Lin^–ckit⁺Sca1⁺CD150⁺CD48^–), short-term hematopoietic stem cells (ST-HSC, Lin^–ckit⁺Sca1⁺CD150⁻CD48^–), multi-potent progenitors (MPP, Lin^–ckit⁺Sca1⁺CD150^–CD48⁺), common lymphoid progenitors (CLP, Lin^–IL7Rα⁺ckit⁺), common myeloid progenitors (CMP, Lin^–ckit⁺Sca1^–FcγR^low/–CD34⁺), megakaryocyte-erythroid progenitors (MEP, Lin^–ckit⁺Sca1^–FcγR^–CD34^–), granulocyte-monocyte progenitors (GMP, Lin⁻ ckit⁺Sca1^–FcγR⁺CD34⁺), bone marrow (BM) myeloid cells (CD11b⁺), total thymocytes, bone marrow (BM) B cells (B220⁺), and peripheral blood red blood cells (RBC) 28 days after BMT in untreated (n = 10), AE-treated (n = 8), VA-treated (n = 10), Ctrl-FMT mice without (n = 10) or with (n = 9) AE-treatment and Res-FMT mice without (n = 10) or with (n = 10) AE-treatment. Presented as percentage in relation to untreated ctrl mice (No FMT no Abx). Results represent two independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001. Data is presented as mean ± SEM. See also Figure S2 and Figure S3.

Fig. 3. Abx-mediated depletion of the intestinal microbiota reduces body fat and caloric uptake from the diet

(A) Weight of untreated (n = 10), AE-treated (n = 8), and VA-treated (n = 10) mice after BMT, relative to day 0. (B) Weight of periovarian fat in untreated (n = 15), AE-treated (n = 14), and VA-treated (n = 10) 28 days after BMT. (C) Representative photographs of intestines and cecum in untreated, AE-treated and VA-treated mice 28 days after BMT. (D) Quantification of weight of intestines + cecum including contents (from duodenum to rectum) of untreated (n = 5), AE-treated (n = 5), and VA-treated (n = 5) mice 28 days after BMT. (E) Energy intake during 24h, (F) energy excreted as feces during 24h, (G) fraction of energy intake absorbed (absorbed energy (ingested calories after subtraction of excreted calories) divided by ingested calories), (H) energy expenditure rate and total energy expenditure during 24h, (I) cumulative and total distance travelled, and (J) respiratory exchange ratio for untreated (n = 10) and AE-treated (n = 10) mice 13 days after BMT. Light and dark cycle indicated by white and grey background, respectively. Results except (C) and (D) represent at least two independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001, n.s. – Not significant. Data is presented as mean ± SEM. See also Figure S4.

Fig. 4. Sucrose supplementation rescues impaired hematopoietic recovery after BMT in mice with a depleted flora

(A) Experimental outline of sucrose supplementation in water and BMT. (B) WBC, RBC, PLT, LYMPH, NEUT, and MONO counts after BMT, (C) total bone marrow cellularity and composition, (D) total thymocyte count, and (E) weight of periovarian fat 28 days after BMT in untreated and AE-treated mice with and without 5% sucrose in drinking water (n = 10 per group). (F) Quantification of bacterial 16S rRNA and (G) principal components 1 and 2 based on weighted normalized Unifrac analysis of 16S Operational Taxonomic Unit (OTU) abundance in fecal samples 28 days after BMT from untreated and AE-treated mice with and without administration of 5% sucrose in drinking water (untreated mice without sucrose administration, n = 3, all other groups n = 5). Numbers within brackets are percent variation explained by the component. Shaded areas in (B) indicate normal ranges. Results except (F) and (G) represent two independent experiments. * P < 0.05, ** P < 0.01, *** P < 0.001, n.s. – Not significant. Data is presented as mean ± SEM. See also Figure S4.

Fig. 5. Dose dependent relationship between intestinal flora injury and post-BMT hematopoiesis

(A) Schematic outline of abx treatment and BMT. (B) Fecal bacterial diversity (Shannon index) and fecal bacterial abundance (16S rRNA copies) in samples collected on day 0 and day 14 after BMT from untreated mice (No abx, n = 6), and mice treated with streptomycin (n = 8), metronidazole (n = 10), aztreonam (n = 6), or ampicillin (n = 4). Ellipses show 95% confidence intervals. (C) Intestinal integrity described as an intestinal microbiota score = Shannon index × fecal bacterial abundance for samples shown in (B). (D) Pearson correlation of intestinal microbiota score 14 days after BMT, weight of periovarian fat, bone marrow-, and thymus cellularity in untreated mice (No abx, n = 3) and mice treated with streptomycin (n = 4), metronidazole (n = 5), aztreonam (n = 4), and ampicillin (n = 3). Results represent one experiment. * P < 0.05, ** P < 0.01. Data is presented as mean ± SEM. See also Figure S5.

Fig. 6. Intestinal flora depletion affects steady-state hematopoiesis

(A) Outline of abx-treatment of mice at steady-state without BMT. (B) Bone marrow cellularity, (C) number of bone marrow B cells (B220⁺), and (D) number of thymocytes in untreated specific pathogen free (SPF) mice (n = 14), AE-treated SPF mice (n = 15), untreated germ-free (GF) mice (n = 10) and AE-treated GF mice (n = 12) after 35 days of antibiotic treatment. (E) Weight of periovarian fat in untreated SPF mice (n = 10) and AE-treated SPF mice (n = 10) after 35 days of antibiotic treatment. Results represent at least two independent experiments. * P < 0.05, *** P < 0.001, n.s. – Not significant. Data is presented as mean ± SEM. See also Figure S6.