Myeloid cells interact with a subset of thyrocytes to promote their migration and follicle formation through NF-κB

The pathogenesis of thyroid dysgenesis (TD) is not well understood. Here, using a combination of single-cell RNA and spatial transcriptome sequencing, we identify a subgroup of NF-κB-activated thyrocytes located at the center of thyroid tissues in postnatal mice, which maintained a partially mesenchymal phenotype. These cells actively protruded out of the thyroid primordium and generated new follicles in zebrafish embryos through continuous tracing. Suppressing NF-κB signaling affected thyrocyte migration and follicle formation, leading to a TD-like phenotype in both mice and zebrafish. Interestingly, during thyroid folliculogenesis, myeloid cells played a crucial role in promoting thyrocyte migration by maintaining close contact and secreting TNF-α. We found that cebpa mutant zebrafish, in which all myeloid cells were depleted, exhibited thyrocyte migration defects. Taken together, our results suggest that myeloid-derived TNF-α-induced NF-κB activation plays a critical role in promoting the migration of vertebrate thyrocytes for follicle generation.

Source data are provided as a Source data file.

Fig. S2 .
Fig. S2.Major cell types identified from thyroid tissues at postnatal day 5, 10, 20, and 30 mice by 10 × scRNA-seq.a, Workflow showing the 10 × single cell sequencing strategy.Mice thyroid tissues at postnatal day 5, 10, 20 and 30 were collected respectively for scRNA-seq.b-e, Sequencing, and quality control metrics for each sample.The distributions maps of each quality control metrics for cells that were retained following removal of low-quality cells.Number of unique genes detected per cell (b), Number of UMIs detected per cell (c), Percentage of UMIs that maps to a mitochondrial protein-encoding gene per cell (d), and UMIs per gene (e) detected in each sample were shown.f-g, Dot plot showing the relationship between the number of UMIs and mitochondrial UMIs (f) or number of genes (g) in each sample after quality control.h, UMAP plot of all cells collected from the mice thyroid tissues at postnatal day 5, 10, 20 and 30.i, Violin plot of expression of canonical marker genes or different

Fig. S3 .
Fig. S3.The expression profile of canonical and novel genes enriched in epithelial cluster identified by 10 × scRNA-seq.a, UMAP plot showing the epithelial cells in thyroid tissues could be clustered into parathyroid cells (PTC), parafollicular cells (C cell) and thyroid follicular cells (TFC).b, The percentage of the three epithelial subtype cells in thyroid tissues.c, The proportion of epithelial subtype cells in total epithelial cells in thyroid tissues at postnatal day 5, 10, 20 and 30 mice.d, Expression pattern of

Fig. S4 .
Fig. S4.Monocle trajectory analysis of the TFC cells.a-c, Monocle2 prediction of the developmental trajectory of the TFC according to pseudo-time (a), cluster information (b) and developmental time (c).d, the DEGs (rows) along the pseudo-time (columns) were clustered hierarchically into two different modules.e, Hallmark gene sets, KEGG and GO analysis of the two gene modules.P value was determined by Benjamini-Hochberg-adjusted one-sided hypergeometric test.Source data are provided as a Source data file.

Fig. S6 .
Fig. S6.The number of thyroid follicular cells with NF-κB activation gradually reduced in mice thyroid tissues from postnatal day 5 to day 30.a-b, Representative images (a) and statistical assessment (b) of P65 positive cells in thyroid tissues at postnatal day 5, 10, 20 and 30 mice by IHC analysis.c, Representative images showing the expression of phosphorylated IKKα/IKKβ in NKX-2-1 positive thyrocytes in the thyroid tissues of postnatal day 5 and day 30 mice by double immunofluorescence assay.d, Statistical assessment of the percentage of thyrocytes with high expression of P-IKKα/IKKβ in thyroid tissues at postnatal day 5, 10, 20 and 30 mice.NKX2-1 as a marker for thyrocyte.e, Representative images showing the expression of phosphorylated P65 (p-P65) in NKX2-1 positive thyrocytes in thyroid tissues at postnatal day 5 and 30 mice by double immunofluorescence assay.f, Statistical assessment of the percentage of nuclear p-P65 positive thyrocytes in thyroid tissues at postnatal day 5, 10, 20 and 30 mice.The number of p-P65 positive thyrocytes in thyroid tissues were gradually reduced with mice growth.Scale bar, 100 μm in a, and 25 μm in c and e.Data are shown as mean ± SD, n = 6 biologically independent samples, statistical significance was determined by One-way ANOVA, followed by Tukey's multiple comparison test (b, d, f).Source data are provided as a Source data file.

Fig. S7 .
Fig. S7.NF-κB activation in thyrocytes during zebrafish folliculogenesis.Continuous observation of NF-κB activation during new follicles formation using zebrafish double transgenic tg:mCherry; nfkb:eGFP line, with time indicated on the bottom.White arrowheads (53.5 hpf, 54 hpf, 54.5 hpf) point to the front extended thyrocytes with NF-κB activation, and white arrowheads (60 hpf and 60.5 hpf) point to the rear migrated thyrocytes with NF-κB activation.From 61 hpf, when the epithelial cluster nearly separated from the TP (circled by white dotted lines) and a new follicle generated, the GFP fluorescent signals in these thyrocytes were faded down.Three independent experiments were carried.Scale bar: 50 μm.

Fig. S8 .
Fig. S8.TPCA1, an IKK2 inhibitor, suppressed NF-κB activation and motility of thyrocytes in mouse thyroid tissues.a-b, Representative image (a) showing the P65 and Pax8 double positive thyrocytes and statistical assessment (b) of the percentage of P65/Pax8 double positive cells in total pax8 positive thyrocytes in thyroid tissues of mice with or without treatment by TPCA1.After treatment with TPCA1 for one month, the number of P65 and Pax8 double positive thyrocytes were dramatically reduced in thyroid tissues at postnatal day 30 mice.c-d, Representative image (c) showing NKX2-1 and MCAM expression by double immunofluorescence assay and statistical assessment (d) of the percentage of NKX2-1/MCAM double positive cells in total NKX2-1 positive thyrocytes in thyroid tissues of mice with or without treatment by TPCA1.e-f, Representative image (e) showing NKX2-1 and Vimentin expression and statistical assessment (f) of the percentage of NKX2-1/Vimentin double positive cells in total NKX2-1 positive thyrocytes in thyroid tissues at postnatal day 30 mice with or without treatment by TPCA1.g-h, Representative image showing multicolor RNA scope staining of Pax8/Map4k4 expression (g) and statistical assessment (h) of the percentage of Pax8/Map4k4 double positive cells among total Pax8 positive thyrocytes in thyroid tissues at postnatal day 30 mice with or without treatment by TPCA1.Scale bar, 50 μm.Data are shown as mean ± SD, n = 6 biologically independent samples, statistical significance was determined by One-way ANOVA, followed by Tukey's multiple comparison test (b, d, f, h).Source data are provided as a Source data file.
Fig. S9.NF-κB activation in thyrocytes is indispensable for zebrafish thyroid folliculogenesis.a, In vivo continuous observation of thyroid folliculogenesis during 65hpf to 76hpf zebrafish embryos treatment by TPCA1 in zebrafish transgenic tg:mCherry line.The thyrocytes extending out of the zebrafish thyroid primordium were suppressed and so the folliculogensis impeded in zebrafish embryos treatment by TPCA1.Three independent experiments carried.b, Representative images showing T4 positive follicles in 8dpf zebrafish embryos with or without TPCA1 treatment by IF staining.c, Representative image, and statistical assessment showing TUNEL positive cell per thyroid gland in 6dpf zebrafish embryos with or without TPCA1 treatment.d, Representative images and statistical assessment of thyroid follicle formed at 6 dpf embryos with or without treated by BMS-345541 using zebrafish transgenic tg:mCherry line.e, Representative images and statistical assessment showing thyroid specific inhibition of NF-κB activation by constitutively overexpression nfkbia under thyroglobulin promoter induced defective thyroid folliculogenesis of 6 dpf embryos using zebrafish transgenic tg:mCherry line.f-g, Comparison of tg expression in 50hpf WT and TPCA1 (f) or BMS (g) treated embryos by WISH.h, Representative image, and statistical assessment of follicles formed in zebrafish embryos treated with NSC23766 or DMSO.i, Representative image and statistical measurement of follicle numbers formed in 6dpf control and TPCA1 treated zebrafish embryos with or without

Fig. S10 .
Fig. S10.The interaction of myeloid cells with thyrocytes identified by cellphoneDB mediated by Tnfα and its receptor pairs.a, Overview of the interaction pairs identified between 14 cell types identified by scRNA-seq.b, Ligand-receptor analysis to assess the number of interaction pairs against TFC cluster split by each time point.c, Numbers of significant interaction pairs between cell types mediated by Tnf.Colors indicate the numbers of significant interaction pairs.The x axis shows the cell type expressing the ligand Tnf and the y axis shows the cell type expressing the receptor.

Fig. S11 .
Fig. S11.Interactions of thyrocyte with myeloid cells promoted the migration of thyroid epithelial cells.a, Myeloid cells are infiltrated into thyroid tissues at postnatal 10 mice.The panels of right two columns (a' and b') are enlarged pictures of areas framed on the left panel, which indicated CD68 positive macrophages infiltrated in the thyroid tissues at postnatal 10 mice.Arrows in (b') pointed to the CD68 positive macrophages located outside of the vascular stained by Col1a1.b, Long time continuous observation of the interaction with macrophages [Tg(mpeg1:eGFP)] promoted the collective migration of epithelial cluster [Tg(tg:mCherry)].During this time window, obvious front epithelial cells migrations were detected, with one (18 minutes to 63 minutes) and another one (177 minutes to 216 minutes), and the start point were marked with white arrowhead and circled.Both epithelial migration events were accompanied with frequent interactions with surrounding macrophages.c, Tnf-α expression in neutrophils by immunofluorescence assays.Scale bar, 50 μm, and three independent experiments were carried for a, b and c.

Fig. S13 .
Fig. S13.Influence of Tnf-α and NF-κB signaling on zebrafish vascular development and mouse folliculogenesis in vitro.a, The influence of mpeg1 positive macrophages surrounding TP with lenalidomide addition.b-c, Double transgenic line shows the vascular Tg(flk:eGFP) growth surrounding TP with lenalidomide, TPCA1 or BMS addition.d, Western blot assessment of migration related protein expression in Nthy-Ori3 normal thyroid epithelial cell line coculture with HL-60 or combined with lenalidomide treatment.e, Three-dimensional Matrigel for assessing folliculogenesis with mouse primary thyroid cells.The second row show little effect of BMS addition on follicle formation.The third panel show increased epithelial motility and muti-lumen formation induced with Tnf-α.Phalloidin for staining cytoskeleton and antibody against