Advertisement

Neutrophils rapidly migrate via lymphatics after Mycobacterium bovis BCG intradermal vaccination and shuttle live bacilli to the draining lymph nodes

Valérie Abadie, Edgar Badell, Patrice Douillard, Danielle Ensergueix, Pieter J. M. Leenen, Myriam Tanguy, Laurence Fiette, Sem Saeland, Brigitte Gicquel and Nathalie Winter

Data supplements

  • Supplemental Figures for: Abadie et al, Vol 106, Issue 5, 1843-1850

    Files in this Data Supplement:

    • Figure S1. BCG is not associated with Langerhans cells in the dermis (JPG, 93 KB) -

      (A) At 12 and 72 hours after BCG-egfp injection, skin explant cells were double-stained with CD11c and anti-CD207 (Langerin) and analyzed by flow cytometry. Among the CD11c+ gated cells, only 4% at 12 hours and less than 1% at 72 hours after injection were CD207+ Langerhans cells. (B) Ear cryosections from BCG-egfp and mock-injected mice were immunostained with anti-CD207. Co-localization of BCG (green) with Langerhans cells (red) could not be observed (magnification, ×200).

    • Figure S2. Langerhans cells do not carry BCG into the ADLN after vaccination in the ear dermis (JPG, 111 KB) -

      (A) At 12 and 72 hours after BCG-egfp injection, ADLN CD11c+ cells were magnetically purified and stained with anti-CD207 before flow cytometry. Less than 1% of Langerhans cells (CD11c+, CD207+) were EGFP+, ie, BCG-infected. (B) ADLN sections from BCG-egfp or mock-injected mice were immunostained with anti-CD207. At 12 and 72 hours after injection, LCs (red) were detected in the paracortex of the ADLN but did not colocalize with BCG (green) that remained in the subcapsular space (magnification, ×100).

Article Figures & Data

Figures

  • Figure 1.

    BCG bacilli are rapidly detected and persist in ADLNs after intradermal vaccination in the ear. BCG (106 CFUs) was inoculated into the dermis of each ear dorsum. From 4 hours to 14 days after injection, each ear (•) and ADLNs (○) were harvested and BCG multiplication was estimated by counting CFUs after plating on selective medium; data are the mean CFUs ± SD from 5 animals, that is, 10 ears and ADLNs processed individually.

  • Figure 2.

    BCG is neither associated with DCs nor with macrophages in the dermis. (A) From 4 to 72 hours after BCG-egfp inoculation, skin explant cells were stained with CD11c and analyzed by flow cytometry. (B) Ear cryosections from BCG-egfp and mock-injected mice were immunostained with antibodies directed against CD11c and mMGL, a receptor largely distributed among dermal phagocytes. Colocalization of BCG (green) with these cell populations (red) could not be observed. After labeling with Texas red (CD11c) or Alexa 594 (mMGL), slides mounted in Fluoromount were analyzed under a Zeiss Axioskop fluorescence microscope with an Achroplan 20 ×/0.45 NA objective. Images were acquired with a Leica DC300F camera and Leica Qwin software. Images were then processed with Adobe Photoshop (Adobe Systems, San Jose, CA).

  • Figure 3.

    DCs do not carry BCG into the ADLNs after vaccination in the ear dermis. (A) Low-density cells recovered from ADLNs from 4 to 72 hours after BCG-egfp injection were stained with anti-CD11c and analyzed by flow cytometry. (B) ADLN sections from BCG or mock-injected mice were immunostained with anti-CD11c. At 72 hours after injection, colocalization of conventional DCs and BCG was observed on rare occasions (inset), but most of the bacilli remained localized under the ADLN capsule. Images were acquired as in Figure 2B, except that cells were labeled only with Texas red, and 10 ×/0.25 NA and 40 ×/0.75 NA (inset only) objectives were used.

  • Figure 4.

    Neutrophils are rapidly recruited to the dermis where they phagocytose BCG and then massively crawl out from the skin. (A) Cells crawling out from ear skin explants from 4 to 72 hours after BCG-egfp vaccination were analyzed by flow cytometry after CD11b staining. (B) Twenty-four hours after vaccination, EGFP+ skin explant cells gated on a side-scatter (SSC-H), EGFP dot plot were phenotyped as CD11b+ Ly-6G+, MHCII-, and F4/80-. (C) CD11b+/EGFP+ cells were sorted by flow cytometry and observed by light microscopy after staining with May-Grünwald-Giemsa. Characteristic polylobed nuclei confirmed that main BCG host cells in skin are neutrophils. In most cases, several bacilli were detected inside each neutrophil (arrows). Cytocentrifuged cells were observed under a Zeiss Axioskop light microscope with an Achroplan 100 ×/1.25 NA oil iris objective. Images were acquired with a Leica 300F camera and processed with Leica Qwin software. (D i-ii, iv-vi) Ear skin cryosections were immunolabeled with anti-Ly-6G (red). Neutrophils that invaded the dermis from 4 to 72 hours after injection were focally organized around bacilli at the injection site. Images were acquired as in Figure 3B. (iii) At 12 hours, bacillus (green) and neutrophil (red) colocalization was observed by confocal microscopy. After labeling with Alexa 594 slides were analyzed under a Zeiss Axioskop 2FS microscope with a Plan-APOCHROMAT 63 ×/1.4 NA objective. Images were acquired and processed with Zeiss LSM 510 software.

  • Figure 5.

    Neutrophils massively infiltrate ADLNs early after vaccination and shelter BCG. (A) At 24 hours after vaccination with BCG-egfp, ADLN EGFP+ gated cells were phenotyped as CD11b+, Ly-6G+, F4/80-, and MHCII-, indicating they were neutrophils. (B) CD11b+ ADLN cells were magnetically enriched and EGFP+ cells were sorted by fluorescence-activated cell sorting (FACS). The majority of the CD11b+/EGFP+ cells were positive for Ly-6G (red), and microscopic analysis revealed their characteristic polylobed nucleus stained with DAPI (blue). Several bacilli (green) were often detected inside the same neutrophil. Cells adhering to slides were labeled with Alexa fluor 594 (red) and DAPI (blue) and observed under a Zeiss Axioskop fluorescence microscope with an Achroplan 100 ×/1.25 NA oil iris objective. Images were acquired with a Leica 300F camera and processed with Leica Qwin software. (C) Neutrophil recruitment to the ADLNs was analyzed at 4, 12, 24, and 72 hours after BCG-egfp injection, by immunostaining cryosections with anti-Ly-6G. From 4 to 24 hours, neutrophils (red) massively infiltrated the subcapsular space and sinuses of the ADLNs and mainly colocalized with BCG (green). Bacilli were detected inside neutrophils under the ADLN capsule as observed by confocal microscopy and indicated by arrows. At 72 hours, neutrophils disappeared from the ADLNs. (i-ii, iv-vi) Images were acquired as in Figure 3B. (iii) Image was acquired as in Figure 4Diii.

  • Figure 6.

    Infected neutrophils leave the skin via afferent lymphatics after BCG injection and shuttle fluorescent bacilli to the ADLNs. (A) A total of 106 CFUs of rBCG-dsred and rBCG-egfp fluorescent strains was injected into 2 adjacent distinct sites of the ear dorsum (left panel). Four hours later, ADLN cryosections were laser scanned under a confocal microscope to colocalize red or green bacilli with Ly-6G+ cells (blue). Neutrophils carrying either red or green bacilli or coinfected with both strains in the ADLNs were scored in 7 fields from 2 ADLN sections. Right panel is control coinjection of mixed red and green bacilli in the same site. Cryosections were labeled with Alexa fluor 633 (blue) mounted in Fluoromount. Otherwise, images were acquired as in Figure 4Diii. (B) Four hours after BCG-egfp inoculation, ear sections were immunostained with anti-Lyve-1 (brown) and neutrophils were detected either by counterstaining polylobed nuclei with hematoxylin or by anti-Ly-6G (blue). Neutrophils were detected inside the lumen of lymphatic vessels in the injection site vicinity. After antibody treatment, paraffin embedded sections were preserved in Aquamount and observed under a Nikon Microphot FXA light microscope with Plan Apo 60 ×/1.40 NA oil iris objective (left) or 40 ×/0.70 NA objective (right). Images were acquired with a Nikon DX digital camera and processed with the Nikon capture software. (C) Four hours following BCG-egfp inoculation, ear dermis was stained with anti-Lyve-1 (red) and anti-Ly-6G (blue), and a 3-dimensional “orthogonal” slice projection was analyzed by confocal microscopy. The large central panel shows a single image among 46 slices recorded at 0.23-μm intervals. To characterize cells inside lymphatic vessels (underlined by red dashes), the x-axis (green line) and y-axis (red line) were defined for sliced z-axis reconstruction. The corresponding results for the x, z slice and y, z slice are shown and the crossing point between green and red lines represents the z-stack position of the central panel image. A neutrophil carrying bacilli inside the lymphatic vessel lumen is depicted. Dermis cryosections were labeled with Alexa fluor 594 (red) and 633 (blue) mounted in Fluoromount under a Zeiss Axioskop 2FS with a Plan-APOCHROMAT 63 ×/1.4 NA objective and zoomed 2.9 ×. Images were acquired and processed with Zeiss LSM 510 software.

  • Figure 7.

    Fluorescent-labeled neutrophils accumulate in the ipsilateral ADLNs and reach the paracortical region. (A) Bone marrow neutrophils were labeled with CFSE (green) or PKH26 (red) and injected inside the left or right ear dorsum concomitantly with 106 CFUs of wild-type BCG (i-ii). Four hours later, green neutrophils were detected only in the left ADLNs and red neutrophils in the right ADLNs (iii-iv). In 3 independent experiments analyzed, no mixed green and red neutrophils were ever observed. Control mock-injection induced only moderate infiltration of fluorescent neutrophils in the regional ADLNs (i-ii). (B) At 12 hours after injection, Ly-6G+ neutrophils (green) were detected inside the T-cell (CD4+ and CD8+ in red) area of the ADLNs (i). CD11c+(red) DCs were also detected in the same zone and contacts between neutrophils and DCs could be observed (ii, arrows). Images were acquired as in Figure 3B.