Zen 2009

For migration imaging, IgA-Cre/YC3.60^flox mice (8–12 weeks-old) were anesthetized with a mixture of three types of anesthetic agents as described previously (19 (link)). An incision was carefully made in the abdominal wall, and the small intestine was exposed. PPs were identified by naked eyes. About 8000 sorted CD11b⁺IgA⁺ PP B cells and 30 000 CD11b⁻IgA⁺ PP B cells were labeled with CellTracker Orange CMTMR fluorescent dye (Invitrogen, USA) and then injected to a PP of an IgA-Cre/YC3.60^flox transgenic mouse directly by a 25 μl syringe (Trajan Scientific and Medical, Australia). The PP with transferred cells was observed under an LSM 880 microscope (Carl Zeiss, Germany). Images were analyzed with ZEN2009 software (Carl Zeiss, Germany). After one-hour observation of the PP under a microscope, the abdominal incision was carefully closed with an ELP Skin Stapler (Akiyama Co. Ltd. Japan). Forty hours after transfer, under anesthesia, the PP with transferred cells was observed again to identify the localization of transferred CD11b⁺IgA⁺ PP B cells under an LSM 880 microscope (Carl Zeiss, Germany) and analyzed with ZEN2009 software (Carl Zeiss, Germany).

All live imaging was performed on an inverted 710 DUO/NLO (Zeiss) microscope with a 40x or 63x water C-apochromat objective, a 7-LIVE scanner with multiple high-speed CCD line detectors, a 488 nm laser, a 561 nm laser, and a 633 nm laser. Zen 2009 (Zeiss) was used for acquisition, ImageJ (NIH) and Volocity (PerkinElmer) were utilized for post-acquisition analysis, and Final Cut Pro (Apple) was used for video editing and preparation.

The imaging zone was maintained to be consistent with that showing the epidermal cells of the root tip meristematic zone, 10–15 cells above the quiescent center. To measure the fluorescence signal in BFA bodies, we obtained 2–4 slices of epidermal cells. Image analysis and signal quantification were performed using the measurement function of the LSM software ZEN 2009 (Zeiss). The signal intensity ratio of a BFA-body region was quantified, normalized to the area, and divided by the signal intensity of a nearby plasma membrane.

Proximity ligation analysis (PLA) was performed according to the manufacturer's protocol (Duolink II Fluorescence, Duolink In Situ Detection Reagents Orange, OLink). Cells were fixed in 4% phosphate-buffered formaldehyde solution, permeabilized with 0.05% Triton X-100, blocked, and incubated overnight with rabbit monoclonal antibodies against NOX4 (provided by one of the co-authors (P.J.-D.)) and CANX antibody (Santa Cruz, number sc-6465) as well as negative and positive controls to validate the assay (data not shown). To show the specificity of the CANX antibody, a blocking peptide was used (Santa Cruz, number sc-6465 P). Samples were washed, incubated with the respective PLA probes for 1 h (37 °C), washed, and ligated for 30 min (37 °C). After an additional washing, amplification with polymerase was performed for 100 min (37 °C). The nuclei were stained with DAPI. Images were acquired by confocal microscope (LSM 510, Zeiss) and Plan-Neofluar ×40/1.3 oil objective at room temperature. Acquisition software Zen 2009 (Zeiss) was used. For each biological sample (n ≥ 3) eight pictures were evaluated and counted. Quantitative analysis was performed using Fiji software. Interactions were normalized to nuclei.

Double immunofluorescence was performed using anti-LRP1 and anti-CD204 antibodies in formalin-fixed paraffin-embedded human ESCC tissue. The nuclei were stained with DAPI (DOJINDO LABORATORIES, Kumamoto, Japan). The lists of primary and secondary antibodies are presented in Table S2. All images were taken with a Zeiss LSM 700 laser-scanning microscope and analyzed using the LSM software ZEN 2009 (Carl Zeiss, Oberkochen, Germany).