Fig 2: Gating strategy for myeloids in spleen samplesFigures 3 and 4 present the flow cytometry results and our gating strategy for myeloids from blood (Figure 3) and spleen (Figure 4) samples, the samples presented in both figures were from CD45.2 positive parabiont. The plots illustrate the sequential gating used to identify myeloid subsets and determine the chimerism rate between CD45.1+ and CD45.2+ parabiotic partners.

Fig 3: tVAT competes with tumour tissue for the infiltration lymphocytes.a, Experimental design of the PAT immune-competition model using C57BL/6J or BALB/c mice with the removal of either PAT or contralateral inguinal adipose tissue (control). b,c, Tumour weights of MC38-OVA (b) or CT26 (c) tumours at day 16 of the experiment in C57BL/6J or BALB/c mice (n = 5). d,e, Flow cytometry analysis of the infiltration of T cells, CD4+ T cells, CD8+ T cells, and tumour-specific CD8+ T cells in MC38-OVA (d) or CT26 (e) tumours, with and without removal of PAT (n = 5). To identify OVA-specific CD8+ T cells, H-2Kb tetramers were employed in the MC38-OVA tumour model that endogenously expresses ovalbumin. f, Representative images of MC38 tumours (left) and tumour weights (right) in C57BL/6J mice following removal of PAT or control, and treatment with ATS-GNP or saline (n = 5). g, Experimental design of the PAT immune-deficient model (BALB/c-Nude or NSG) with removal of PAT. h,i, Representative images of MC38 tumours (left) and tumour weights (right) at day 16 in C57BL/6 J, BALB/c-Nude (h), or NSG mice (i) (n = 6). j, Schematic representation of the strategy to detect CD45.1+ cells in MC38-OVA tumours, with and without removal of PAT or contralateral inguinal adipose tissue (control), using the CD45.2 receptor mouse model transferred with CD45.1 splenocytes from CD45.1 donor mice. i.v., intravenous. k, Quantitative analysis of various types of CD45.1+ cells in MC38-OVA tumours with and without removal of PAT or control, as determined by flow cytometry (n = 5). Data represent ≥3 independent experiments. Statistical significance was assessed using a two-sided, unpaired Student’s t-test (d,e,k), one-way analysis of variance (ANOVA) with Tukey’s correction for multiple comparisons (b,c,f right), a two-way ANOVA with Tukey’s correction for multiple comparisons (h right) or a two-way ANOVA with Tukey’s correction for multiple comparisons (i right). Graphs display mean ± s.d. (b–f,h i,k). Panels created with BioRender: a, g and j, Huaiqiang, J. https://biorender.com/f9uacgq (2026).Source data

Fig 4: Pld4 deficiency in mice results in autoimmunity and cell-intrinsic expansion of pDCs and plasma cells.a,b, The autoimmune phenotypes in Pld4-deficient mice. a, Body-weight change between Pld4−/− and wild-type mice. n = 11 (WT), n = 11 (Pld4+/−) and n = 11 (Pld4−/−) mice. The asterisks represent statistical comparisons between Pld4−/− and wild-type mice. b, Plasma anti-dsDNA and anti-dsRNA antibodies levels of Pld4−/− and wild-type mice. Female: n = 10 (WT), n = 10 (Pld4+/−), n = 17 (Pld4−/−); male: n = 8 (WT), n = 12 (Pld4+/−) and n = 10 (Pld4−/−) mice. OD450, optical density at 450 nm. c, UMAP visualization of type I IFN pathway genes in the kidney scRNA-seq data of Pld4−/− and wild-type mice. d–g, The kidney flow cytometry results of Pld4−/− and wild-type mice: immune cells (CD45+) (d), pDCs (CD45+CD3−CD19−CD11b−B220+CD11c+Siglec-H+BST2+) (e), plasma cells (PCs) (CD45+CD3−CD11b−B220low/−CD138+CD98+) (f) and CD4 effector T cells (CD45+CD3+CD4+CD44+CD62L−) (g). n = 12 (WT), n = 12 (Pld4+/−) and n = 12 (Pld4−/−) mice. h,i, The results of mixed bone marrow chimeric mice reconstituted with 1:1 ratio of CD45.1-WT–CD45.2-WT (+/+:+/+) or CD45.1-WT–CD45.2-Pld4−/− (+/+:−/−) bone marrow. h, The plasma anti-dsDNA and anti-dsRNA antibodies levels of chimeric mice. n = 20 (+/+:+/+) and n = 26 (+/+:−/−). i, Kidney immune cell phenotyping of chimeric mice. n = 15 (+/+:+/+) and n = 15 (+/+:−/−). Data are mean ± s.e.m. The results are representative of at least three independent experiments (b and d–g), two independent experiments (a) and a summary of two independent experiments (h and i). Statistical analysis was performed using two-way ANOVA (a and i), one-way ANOVA with Tukey’s post hoc analysis (b and d–g) and unpaired t-tests (h).Source data

Fig 5: CS-induced TRM cells derived from circulating lymphocyte recruitment and in situ proliferation.a The scheme indicated the time points of FTY720 treatment (20 μg in vehicle per time). FTY720 treatment from 4 weeks, in which circumstance, lymphocytes could be recruited into the lung at the inflammatory stage but blocked at the fibrogenesis stage. b H&E staining to the lung section of different treated mice. Scale Bar = 500 μm (upper) and 200 μm (lower). c The inflammation scores were assessed in the lung sections. d FC analysis of CD4+ TRM cells in the lungs of distinct FTY720-challenge. The bar graph shows percentages and counts of CD4+ TRM cells. e Flow histogram indicated Ki-67 expression in CD4+ TRM cells. f FC analysis of CD4+ TRM cells for CD69 and CD103 expressions. The bar graph displayed the ratios of CD103– to CD103+ in the CD69+ TRM cells. g FC analysis of Tregs (FOXP3+) in the CD4+ TRM cells. The bar graph shows the percentages of TRM-Tregs. h Schematic overview of parabiosis experiment. CD45.1/1 and CD45.1/2 mice were approximated with sutures. With 14 days’ recovery, CD45.1/1 mice were treated by CS particle instillation and analyzed 7 days later. i Typical FC plot of circulating blood leukocytes of the parabiont. j Flow plot analysis of the CD4+ TRM cells and circulating TEM cells of the CS-treated conjoined mice. CD45.2+ TRM cells indicated the recruited cells from circulating. k Typical flow plots showed CD69 and CD103 expressing patterns on TRM-Teff cells or TRM-Tregs. Flow plot analysis of CD45.2+ cells in CD69+CD103+ Tregs, CD69+CD103+ Teffs, CD69+CD103- Tregs, and CD69+CD103- Teffs in CD4+ TRM cells. For panels a–g, n = 5 biologically independent animals, n number also indicated independent experimental replicates. Individual mice are plotted on the graphs. Values are reported as the mean ± SD. P value was determined using one-way ANOVA followed by Tukey’s test. For panels j and k, n = 4 biologically independent animals, n number also indicated independent experimental replicates. CD45.1/1 and CD45.1/2 mice approximated with sutures were used in these experiments. Values are reported as the mean ± SD. P value was determined by paired two-tailed Student’s t-test.