AbstractFibroblastic reticular cells (FRCs) are immunologically specialized myofibroblasts of lymphoid organ, and FRC maturation is essential for structural and functional properties of lymph nodes (LNs). Here we show that YAP and TAZ (YAP/TAZ), the final effectors of Hippo signaling, regulate FRC commitment and maturation. Selective depletion of YAP/TAZ in FRCs impairs FRC growth and differentiation and compromises the structural organization of LNs, whereas hyperactivation of YAP/TAZ enhances myofibroblastic characteristics of FRCs and aggravates LN fibrosis. Mechanistically, the interaction between YAP/TAZ and p52 promotes chemokine expression that is required for commitment of FRC lineage prior to lymphotoxin-尾 receptor (LT尾R) engagement, whereas LT尾R activation suppresses YAP/TAZ activity for FRC maturation. Our findings thus present YAP/TAZ as critical regulators of commitment and maturation of FRCs, and hold promise for better understanding of FRC-mediated pathophysiologic processes. IntroductionA fine network of fibroblastic reticular cells (FRCs) is essential for maintaining lymph node (LN) structure and function1,2,3. FRCs regulate immune cell entry into the LN via high endothelial venules (HEVs) and confer compartmentalization of lymphocytes within the LN by secretion of essential chemokines1,4,5,6. In this regard, fibrotic damage to FRCs by chronic inflammation and cancer has been shown to seriously deteriorate systemic immune responses by FRCs7,8,9,10. Thus, proper differentiation of FRC progenitors into mature FRCs during development is critical for acquisition of immunoregulatory characters and initiation of adaptive immune response and chemokine production1,11,12,13,14,15,16.The differentiation of FRCs involves differentiation of a poorly defined population of mesenchymal cells into FRC precursors, which further develop into mature FRCs1,12. Whereas the molecular details involved in the latter process such as lymphotoxin-尾 receptor (LT尾R) and receptor activation of NF-kB ligand (RANKL)-mediated interactions of lymphoid tissue inducer (LTi) cells with FRC precursors have been thoroughly described17,18, characterization of stromal cells and signaling pathways involved in the commitment steps of FRC development are incompletely defined19,20,21.The core of the Hippo pathway consists of large tumor suppressors 1 and 2 (LATS1/2), and yes-associated protein (YAP) and transcriptional co-activator with PDZ-binding motif (TAZ), which are final effectors of the Hippo pathway that exert their functions by mainly interacting with the TEAD/TEF family of transcription factors22,23. Upon activation of Hippo pathway, LATS1/2 become phosphorylated and inhibit the activities of YAP and TAZ (YAP/TAZ). This pathway acts as a key regulator of cellular proliferation, differentiation, organ size control, tissue homeostasis, and regeneration22,24,25,26. Despite such diverse and important roles of Hippo pathway in a wide range of biological processes, its role is largely unexplored in LN FRCs.We hypothesize that the Hippo pathway plays crucial roles in regulating differentiation of FRCs, with relevance to structural and functional properties of LNs. Using FRC-specific genetic targeting and lineage-tracing approaches, we show that YAP/TAZ deficiency impairs FRC differentiation, while hyperactivation of YAZ/TAZ induces myofibroblastic FRCs and LN fibrosis. Thus, we present YAP/TAZ as critical regulators in maintaining FRC integrity and hold promise for better understanding of FRC-mediated physiologic and pathologic conditions.ResultsYAP/TAZ support growth and structure of LN by FRCsTo gain insights into the role of the Hippo pathway in LNs, we first examined the expressions and distributions of YAP/TAZ in human and mouse LNs. Both YAP/TAZ were enriched in 伪-smooth muscle actin (伪SMA)+ FRCs of healthy human LNs (Supplementary Fig.1a). Similarly, YAP/TAZ were highly and equally distributed in nucleus and cytoplasm of FRCs, in addition to endothelial cells of high endothelial venules (HEVs) in mouse LNs (Supplementary Fig.1b).To elucidate the roles of YAP/TAZ in LN FRCs during development, we generated Yap/Taz鈭咶RC mice by crossing Ccl19-Cre mouse20 and Yapflox/flox27/Tazflox/flox28 mouse and analyzed them at 8 weeks after birth (Fig.1a and Supplementary Fig.1c). Cre-negative but flox/flox-positive mice among the littermates were defined as wild-type (WT) mice for each experiment. Although we confirmed the depletion of YAP/TAZ in FRCs of Yap/Taz鈭咶RC mice, there was no difference in body growth except for a slight reduction in LN weight (Fig.1b鈥揹). Of note, LNs of Yap/Taz鈭咶RC mice revealed reduced total cell number (~64.9%), decreased proportion (~42.9%) and number (~61.1%) of FRCs among CD45鈭?/sup> stromal cells, which were less proliferative (~65.2%) but had no difference in apoptosis (Fig.1d鈥揼 and Supplementary Fig.1d-f). However, LNs of Yap/Taz鈭咶RC mice contained similar numbers of BECs and LECs compared with WT (Fig.1g). To address the selective role of YAP or TAZ in FRCs, we compared LNs of Yap螖FRC or Taz螖FRC mice with those of WT and Yap/Taz螖FRC mice (Supplemental Fig. 2a). Of note, decreased LN weight and reduced cellularity observed in Yap/Taz螖FRC mice were not evident in Yap螖FRC or Taz螖FRC mice (Supplemental Fig.2b), suggesting largely redundant roles of YAP and TAZ in FRCs.Fig. 1: YAP/TAZ support growth and structural organization of LNs by FRCs.a Diagram for generation of indicated mice and their analyses at 8 weeks after birth. b, c Representative images of YAP or TAZ in PDGFR尾+ or CCL19+ FRCs in WT and Yap/Taz鈭咶RC mice. FRCs around high endothelial venule (HEV) within the white dashed-line box are magnified in the lower panels with single-channel YAP or TAZ image. Scale bars, 250鈥壜祄. d Comparisons of body weight, inguinal LN weight and total number of cells within the inguinal LN in WT (n鈥?鈥?1; body weight) and Yap/Taz鈭咶RC mice (n鈥?鈥?; body weight). e Representative flow cytometric analysis and comparison of proportion of PDPN+CD31鈭?/sup>FRCs (red box) gated from CD45鈭?/sup> stromal cells of skin-draining LNs in WT and Yap/Taz鈭咶RC mice. f Representative images and comparison of Ki-67+ FRCs (white arrows) in WT and Yap/Taz鈭咶RC mice. Scale bars, 50鈥壜祄. g Comparison of indicated stromal cell counts gated from CD45鈭?/sup> cells of skin-draining LNs in WT and Yap/Taz鈭咶RC mice. BECs (n鈥?鈥?), blood endothelial cells; LECs (n鈥?鈥?), lymphatic endothelial cells. h Representative images of distinction between B and T cells (white dashed line) beneath the LN capsule (white line) in WT and Yap/Taz鈭咶RC mice. Scale bars, 200鈥壜祄. i Comparison of indicated mRNA expression in FRCs sorted from WT and Yap/Taz鈭咶RC mice (quintuplicate values using n鈥?i>=鈥?0鈥?5 mice/group). j, k Representative images and comparison of DsRed+ B cells and GFP+ T cells within the inguinal LN at 24鈥塰 after the adoptive transfer in WT and Yap/Taz鈭咶RC mice. Scale bars, 500鈥壜祄. l Changes in body weight after 1鈥壝椻€?03鈥塸fu of A/PR/8 influenza viral infection (n鈥?鈥?3). m Flow cytometric analyses and comparisons of IFN-纬+CD8+ T cells in gated CD3蔚+ T cells. n鈥?鈥? (CO) or 7 (IM) mice. Unless otherwise denoted, each dot indicates a value obtained from one mouse and n鈥?鈥? mice/group pooled from two independent experiments. Horizontal bars indicate mean鈥壜扁€塖D and P values versus WT by two鈥恡ailed Mann鈥揥hitney U test. NS, not significant.Full size imageFurther analysis of LNs revealed that the distinct border between B and T cell zones was disrupted in Yap/Taz鈭咶RC mice, but not in Yap螖FRC or Taz螖FRC mice (Fig.1h and Supplementary Fig.2c). However, neither proliferation nor apoptosis of immune cells was altered (Supplementary Fig.3a,b). Instead, analysis of isolated FRCs revealed that mRNA levels of lymphoid chemokines for immune cell trafficking were significantly attenuated in Yap/Taz螖FRC mice compared with WT (Fig.1i and Supplementary Fig.3c). Indeed, injection of labeled cells showed that the recruitment of transferred GFP+ T cells (~63.5%) and DsRed+ B cells (~57.0%) were significantly impaired in LNs of Yap/Taz鈭咶RC mice compared with WT mice (Fig.1j, k).Next, we assessed whether the antiviral immune responses would be affected as a consequence of the changes in Yap/Taz鈭咶RC mice by intranasal inoculation of influenza virus (A/PR/8/34). Although there were no apparent differences in immune cell composition in bone marrow, thymus, peripheral circulating blood, and LNs (Supplementary Fig.4a-i), Yap/Taz鈭咶RC mice had less activated interferon-纬 (IFN-纬)-secreting effector CD8+ T cells, and were severely affected compared with WT mice after challenge with influenza virus (Fig.1l,m). However, no significant differences were observed in IL-2+ CD4+ helper T cells and anti-A/PR/8 IgG antibody production between WT and Yap/Taz鈭咶RC mice (Supplementary Fig.5a-c). To better characterize the changes in the FRC phenotype upon Yap/Taz deletion, we generated WTFRC-TR and Yap/Taz鈭咶RC-TR mice by crossing Yap/Taz鈭咶RC mouse with Rosa26-tdTomato reporter mouse (Supplementary Fig.6a,b). LNs of Yap/Taz鈭咶RC-TR mice revealed increased FRC surface area (~1.5-fold) and ER-TR7+ naked conduits (~5.2-fold) compared with WTFRC-TR (Supplementary Fig.6c,d). In concordance, electron microscope revealed that ~45.8% of conduits were not covered with FRCs and ~30.4% of collagen fibrils were irregularly distributed in the remaining conduits in LNs of Yap/Taz鈭咶RC mice compared with WT (Supplementary Fig.6e,f). However, conduit integrity and functionality were relatively preserved despite the reduced FRC coverage in Yap/Taz鈭咶RC mice compared with WT (Supplementary Fig.6g,h), suggesting that minor structural defects in LN conduit system are unlikely the main cause of altered immune cell trafficking.YAP/TAZ hyperactivation impairs differentiation of FRCsTo specifically hyperactivate YAP/TAZ in FRCs during LN development, we generated Lats1/2鈭咶RC mutants by crossing Ccl19-Cre mouse and Lats1flox/flox鈥?sup>29/Lats2flox/flox鈥?sup>30 mouse (Fig.2a). Neonatal Lats1/2鈭咶RC mice exhibited substantial growth retardation with lethality at 16鈥?1 days after birth (Supplementary Fig.7a-c). Moreover, they showed severely disrupted structural organization of LNs with impaired lymphatic drainage (Fig.2b and Supplementary Fig.7d,e). LN weight (~61%), total number of cells within the LN (~98%), and the fraction of stromal cell populations of LNs were also markedly reduced in Lats1/2鈭咶RC mice compared with WT (Fig.2c and Supplementary Fig.7f).Fig. 2: YAP/TAZ hyperactivation impairs differentiation and maturation of FRCs.a Diagram for analyses of indicated mice at P14. b Representative images of PDGFR尾+ FRCs and CD31+ vessels in WT and Lats1/2鈭咶RC mice (n鈥?鈥?). Scale bars, 500鈥壜祄. c Comparisons of LN weight (n鈥?鈥?鈥?) and total number of cells (n鈥?鈥?鈥?0) in WT and Lats1/2鈭咶RC mice. d Diagram for analyses of indicated mice at E18.5 or P14. e Representative images of LN anlagen (dashed line) at E18.5 showing CD4+ LTi cells in WT and Lats1/2螖FRC mice (n鈥?鈥?). Scale bars, 200鈥壩糾. f, Representative images of indicated markers (dashed box) within the inguinal LN (dotted-line) in WT and Lats1/2螖FRC mice at P14 (n鈥?鈥?). Scale bars, 500鈥壜祄. g, h Diagram and representative images for analyses of WT螖FRC-TR mice (n鈥?鈥?) that were injected with anti-CD3蔚 for 5 days to induce T cell depletion. Scale bars, 100鈥壩糾. i Representative flow cytometric plots and comparison of proportion of PDPN+CD31鈭?/sup> FRCs (red box) and PDPN鈭?/sup>CD31鈭?/sup> double-negative (DN) cells of skin-draining LNs in WT and Lats1/2鈭咶RC (n鈥?鈥?鈥?) mice. j Representative images and comparison of YAP expression and nuclear localization (green-arrowheads) in LN of WT and Lats1/2鈭咶RC mice (n鈥?鈥?). Scale bars, 20鈥壜祄. k Comparison of indicated mRNA expression in FRCs sorted from WT螖FRC-TR and Lats1/2i螖FRC-TR mice (n鈥?鈥?). l Representative images and comparisons of indicated marker expressions in LNs of WT and Lats1/2鈭咶RC mice (n鈥?鈥?鈥?). Scale bars, 20鈥壜祄. m Comparison of indicated mRNA expression in FRCs sorted from WT螖FRC-TR and Lats1/2i螖FRC-TR mice. n Diagram for analyses of indicated mice at P14. o Representative images of YAP expression in LNs of WT and L1/2-Y/T鈭咶RCmice. Scale bars, 100鈥壜祄. p, q Representative images of indicated markers in LNs of WT and L1/2-Y/T鈭咶RC mice. Scale bars, 500鈥壜祄. Unless otherwise denoted, each dot indicates a value obtained from inguinal LN and n鈥?鈥? mice. Horizontal bars indicate mean鈥壜扁€塖D and P values versus WT or WT螖FRC-TR by two鈥恡ailed Mann鈥怶hitney U test except for (k) and (m) (two-tailed Student鈥檚 t-test). NS, not significant.Full size imageSince immune cells reorganize the reticular network after birth3, we examined the recruitment of lymphoid tissue inducer (LTi) cells and the organization of lymphoid tissue organizer (LTo) cells both at the embryonic and postnatal period in Lats1/2鈭咶RC mice (Fig.2d).At embryonic (E) day 14.5, engagement of LECs and CD4+ LTi cells were observed in WT and Lats1/2螖FRC embryos (Supplementary Fig.7g), which led to preserved recruitment of LTi cells at E18.5 (Fig.2e). Nevertheless, substantial defects in organization of LTo cells were observed at E18.5 and postnatal day (P)5 in both Yap/Taz鈭咶RC and Lats1/2鈭咶RC mice compared with WT mice (Supplemental Fig.7h-j). Thus, adequate activation of YAP/TAZ in FRCs is required for LN development. Moreover, the distinct border between B and T cell zones was disrupted in both Yap/Taz鈭咶RC and Lats1/2鈭咶RC mice compared with WT mice at P7 (Supplemental Fig.7k), whereas immune cells were rarely observed in growing LNs of Lats1/2鈭咶RC neonates compared with WT neonates at P14, implying that immune cell recruitment was impaired during the postnatal period in Lats1/2螖FRC mice (Fig.2f). To evaluate whether the apparent dense distribution of FRCs was due to the reduced number of immune cells, we depleted T lymphocytes31 and other immune cells32 by daily injection of anti-CD3蔚 mAb into WT鈭咶RC-TR mice (Fig.2g). Of note, while T cells were depleted, increase in FRC density was evident in anti-CD3蔚 mAb-injected mice compared with control mice (Fig.2h), indicating that impaired immune cell recruitment could conversely affect the density of FRCs during development.Further analysis of LNs in Lats1/2鈭咶RC mice at P14 showed a reduction in FRCs (~52%) but an increase in PDPN鈭?/sup>/CD31鈭?/sup> double-negative (DN) populations (~1.8-fold) compared with WT mice (Fig.2i). To investigate whether LN phenotypes of Lats1/2鈭咶RC mice are caused by impaired differentiation of FRCs, we performed lineage-tracing assay using Lats1/2鈭咶RC-TR mice, generated by crossing Lats1/2鈭咶RC mouse with Rosa26-tdTomato reporter mouse (Fig.2a). We confirmed enhanced nuclear localization of YAP with upregulated YAP target genes in LN FRCs of Lats1/2鈭咶RC mice and Lats1/2鈭咶RC-TR mice compared with their controls (Fig.2j, k). Indeed, we found high expression of Tomato in PDGFR尾+ cells within the LNs of Lats1/2鈭咶RC-TR mice, indicating that these cells arose from FRC precursors (Supplementary Fig.7l). In addition, Tomato+ FRCs of Lats1/2鈭咶RC-TR mice highly expressed 伪SMA, collagen IV, and PDGFR尾, which are canonical myofibroblastic markers of FRCs but also direct targets of YAP/TAZ33 (Fig.2l). In contrast, expressions of markers of differentiated FRCs including CCL19 and CCL21 were markedly attenuated in FRCs of Lats1/2鈭咶RC-TR mice compared with WT mice (Fig.2l, m), suggesting that YAP/TAZ hyperactivation impairs FRC differentiation during development.Because LATS1/2 can target several pathways25, we sought to ascertain if YAP/TAZ are indeed the pivotal target responsible for the aforementioned phenotypes. In this regard, we generated L1/2-Y/T鈭咶RC mice by crossing Lats1/2鈭咶RC and Yap/Taz鈭咶RC mice (Fig.2n, o). Of note, the vessels within the LN and the distinct B and T cell zones were partially restored in L1/2-Y/T鈭咶RC mice compared with WT mice (Fig.2p, q), implying that YAP/TAZ are major targets of LATS1/2 in FRC differentiation during LN development.YAP/TAZ are dispensable in adult mature FRCsTo uncover the roles of YAP/TAZ in adult mature LN FRCs, we generated i-Yap/Taz鈭咶RC mice by crossing Pdgfrb-CreERT2 mouse, for which we confirmed high Cre activity (~85%) in PDGFR尾+ FRCs (Supplementary Fig.8a-c), with Yapflox/flox/Tazflox/flox mouse and administered tamoxifen to 4-weeksold mice and analyzed them after 4 weeks (Fig.3a). No apparent differences were observed in weight, cellularity, chemokine expression, border of B and T cell zones and distribution of lymphatic vessels in inguinal LNs of Yap/Tazi鈭咶RC mice compared with WT mice (Fig.3b鈥揹 and Supplementary Fig.8d), implying that YAP/TAZ are dispensable for adult mature LN FRCs.Fig. 3: Canonical Hippo pathway LATS1/2-YAP/TAZ governs FRCs.a Diagram for generation of indicated mice and their analyses at 8-weeks old after the tamoxifen injection from 4-weeks old. b Comparisons of the inguinal LN weight and cellularity within the inguinal LN in i-WT螖FRC-TR and i-Yap/Taz螖FRC-TR mice. c Representative images of intact border between B and T cell zones (white dashed line) beneath the LN capsule (white line) in i-WT螖FRC-TR and i-Yap/Taz螖FRC-TR mice (n鈥?鈥?). Scale bars, 200鈥壩糾. d Representative images of preserved LYVE-1+ lymphatic vessels and CD31+ blood vessels within the inguinal LN in i-WT螖FRC-TR and i-Yap/Taz螖FRC-TR mice (n鈥?鈥?). The regions within the white dashed-line box around subcapsular sinuses (SCS), medullary sinus (MS) and HEVs are magnified as indicated. Scale bars, 500鈥壩糾. e Diagram for generation of indicated mice for their analyses at 8-weeks old after the tamoxifen delivery from 6-weeks old. f Comparisons of the inguinal LN weight and total number of cells within the inguinal LN in WT, i-Lats1/2鈭咶RC or i-L1/2-Y/T鈭咶RC mice. g Representative images of inguinal LN in WT, i-Lats1/2鈭咶RC or i-L1/2-Y/T鈭咶RC mice. Scale bars, 500鈥壩糾. h, i Representative images and comparison of YAP nuclear localization (white arrowheads) in inguinal LN of WT, i-Lats1/2鈭咶RC or i-L1/2-Y/T鈭咶RC mice. Scale bars, 40鈥壜祄. j, k Representative images and comparisons of indicated marker expressions in FRCs around T cell zone of inguinal LN in WT, i-Lats1/2鈭咶RC or i-L1/2-Y/T鈭咶RC mice. Scale bars, 60鈥壜祄. l Heatmap and hierarchical clustering of differentially expressed genes of RNA-Seq data in isolated FRCs from WT and i-Lats1/2鈭咶RC mice and list of selected downregulated genes (green) encoding cytokines and chemokines and upregulated genes (red) involved in TGF-尾 signaling. m Canonical IPA-annotated pathways listed in absolute IPA activation Z-score (P鈥?lt;鈥?.05) to identify potential activation or inhibition of indicated signaling pathways in isolated FRCs from i-Lats1/2鈭咶RC mice compared with WT. Unless otherwise denoted, each dot indicates a value obtained from one mouse and n鈥?鈥? mice/group pooled from two independent experiments. Horizontal bars indicate mean鈥壜扁€塖D and P values versus WT, i-WT螖FRC-TR or i-Lats1/2鈭咶RC by two鈥恡ailed Mann鈥怶hitney U test. NS, not significant.Full size imageTo examine whether the canonical LATS1/2-YAP/TAZ pathway also operates during adulthood, we generated i-L1/2-Y/T鈭咶RC mice by crossing Lats1flox/flox/Lats2flox/flox mouse with i-Yap/Taz鈭咶RC mouse, and administered tamoxifen to 6-weeks-old mice and analyzed them 2 weeks later (Fig.3e). Of note, size, weight, cellularity and distribution of PDGFR尾+ FRCs in inguinal LNs of L1/2-Y/Ti鈭咶RC mice were comparable to those of WT mice (Fig.3f, g). Furthermore, although expressions of 伪SMA, collagen IV, PDGFR尾, and PDPN in LN FRCs were increased in i-Lats1/2鈭咶RC mice, those in i-L1/2-Y/T鈭咶RC mice were comparable to WT mice (Fig.3h鈥搆). In support of this notion, our ATAC-sequencing analysis showed enrichment of Pdpn promoter region upon Lats1/2 deletion, which, on the other hand, was abrogated upon Yap/Taz deletion (Supplementary Fig.9a). These results indicate that YAP/TAZ activation must be controlled to maintain the homeostasis of LNs during adulthood.Identification of YAP/TAZ-regulated pathways in FRCsTo gain insights into roles of YAP/TAZ in FRC differentiation and maturation, we performed transcriptomic analysis of gene expression profiles in isolated FRCs from LNs of WT and Yap/Taz鈭咶RC mice. Gene Set Enrichment Analysis (GSEA) disclosed significant reductions in genes regulating the epithelial-mesenchymal transition (EMT) and E2F target genes together with reductions in YAP target genes in FRCs of Yap/Taz鈭咶RC compared with WT (Supplementary Fig.9b,c). Further analysis of FRCs from LNs of i-Lats1/2鈭咶RC mice showed that genes related to EMT and TGF-尾 signaling were enriched together with upregulation of YAP targets compared with WT mice (Supplementary Fig.9d,e). Consistently, TGF-尾 signaling-associated genes such as Bmp4, Bmp3, Tgfb2, and Ogn and fibrosis-associated genes including Wisp2, Nov, Fgf18, Ctgf, and Pdgfa were ranked among the top 10 upregulators by YAP/TAZ hyperactivation in FRCs (Fig.3l and Supplementary Fig.9f). Indeed, genes related with extracellular matrix (ECM), Rho signaling and actin-cytoskeleton rearrangement were upregulated, while those encoding cytokines and chemokines including Ccl19, Ccl21, Il-7, and Baff were downregulated in FRCs of i-Lats1/2鈭咶RC mice compared with WT mice (Fig.3l, m and Supplementary Fig.9g). Ingenuity Pathway Analysis (IPA) revealed that the signaling pathways related to both canonical and non-canonical NF-魏B signaling and production of chemokines were downregulated in LN FRCs of Lats1/2i鈭咶RC mice compared with WT mice (Fig.3m).YAP/TAZ hyperactivation in FRCs impedes immune responseTo assess whether defects in FRC differentiation influence adaptive immune response in i-Lats1/2鈭咶RC mice, we first examined the expressions of homeostatic chemokines and lymphocyte survival factors in FRCs6,20. Lower mRNA levels of Ccl19, Ccl21, Il-7, and Baff in addition to lower levels of CCL19 and CCL21 were found in LN FRCs of i-Lats1/2鈭咶RC mice compared with WT mice (Supplementary Fig.10a-d). Although the structures of HEVs were preserved, homing of labeled-lymphocytes into the LNs of i-Lats1/2鈭咶RC mice was markedly impaired after the adoptive transfer (Supplementary Fig.10e-i). When we performed adoptive transfer with CFSE-labeled OT-II CD4+ T cells followed by immunization with ovalbumin (OVA) via footpad injection, both proliferation and activation of OT-II CD4+ T cells were impaired at 3 days after the OVA injection in LNs of i-Lats1/2鈭咶RC mice compared with those of WT mice (Supplementary Fig.10j-l). Nevertheless, neither alteration in production or differentiation of lymphocytes in bone marrow nor proportion of immune cells including T and B lymphocytes in LNs were found (Supplementary Fig.11a-d). Thus, defects in FRC differentiation by YAP/TAZ hyperactivation primarily results in impaired adaptive immune response.YAP/TAZ are activated in myofibroblastic FRC precursorsOf note, LT尾R signaling, which is known to regulate FRC maturation20,21, was predicted to be preferentially and significantly influenced by YAP/TAZ hyperactivation in FRCs by our IPA analysis (Fig.3m). We therefore sought to corroborate the interaction between LT尾R and YAP/TAZ in FRCs by generating LTbR鈭咶RC-YR mice by crossing Ccl19-Cre mouse with Ltbrflox/flox and YFP reporter mouse (Fig.4a). Consistent with previous reports20,21, levels of 伪SMA, PDGFR尾 and collagen IV in FRCs were increased, while levels of LT尾R, CCL19 and CCL21 were decreased in Ltbr鈭咶RC mice compared with those of WT mice (Fig.4b). Importantly, YAP/TAZ were nuclear localized and expressions of YAP target genes were upregulated in FRCs of Ltbr鈭咶RC mice compared with WT mice (Fig.4c, d), suggesting that YAP/TAZ activity could be negatively regulated by LT尾R signaling.Fig. 4: FRC-specific depletion of Ltbr activates YAP/TAZ-induced myofibrosis.a Diagram for generation of indicated mice and their analyses at 8-weeksold. b Representative images and comparisons of indicated marker expressions on CCL19-YFP+ FRCs in WT鈭咶RC-YR and Ltbr鈭咶RC-YR mice. Scale bars, 20鈥壜祄. c Representative images and comparisons of YAP and TAZ nuclear localization (white arrows) in inguinal LN of WT鈭咶RC and Ltbr鈭咶RC mice. Scale bars, 20鈥壜祄. d Comparison of indicated mRNA expression in FRCs sorted from WT鈭咶RC and Ltbr鈭咶RC mice. Each dot indicates a mean of quadruplicate values using n鈥?i>=鈥?鈥?2 mice/group from three independent experiments. e Diagram for primary culture of FRCs derived from i-Lats1/2鈭咶RC-TR mice and treatment with EtOH (control) or 4-OHT at 4 days after the culture and their analyses at 2 days after the treatment. f Immunoblot analysis of indicated proteins in primary cultured mouse FRCs after treatment with EtOH or 4-OHT for 2 days. g Comparisons of indicated mRNA expression normalized to Gapdh in primary cultured mouse FRCs after treatment with EtOH or 4-OHT for 2 days. Each dot indicates a mean of triplicate values from three independent experiments. h, i Representative images and comparisons of indicated marker expressions in primary cultured mouse FRCs after treatment with EtOH or 4-OHT for 2 days. Scale bars, 30鈥壩糾. Each dot indicates a mean of triplicate values from three independent experiments. j Diagram for primary culture of human FRCs for 4 days and infection with an adenovirus to induce overexpression of active YAP (YAP5SA) or TAZ (TAZ4SA) for their analyses at 2 days after the infection. k, l Representative images and comparisons of indicated marker expressions in primary cultured human FRCs infected with control-, YAP5SA-, or TAZ4SA-adenovirus. Scale bars, 30鈥壩糾. Each dot indicates a mean of triplicate values from three independent experiments. Unless otherwise denoted, each dot indicates a value obtained from one mouse and n鈥?鈥? mice/group pooled from two independent experiments. Horizontal bars indicate mean鈥壜扁€塖D and P values versus WT鈭咶RC or WT鈭咶RC-YR by two鈥恡ailed Mann鈥怶hitney U test except for (g), (i), and (l) (two-tailed Student鈥檚 t-test). NS, not significant.Full size imageTo evaluate whether YAP/TAZ hyperactivation-induced myofibroblastic phenotypes in vivo are mediated through FRC-intrinsic signaling, we cultured primary FRCs derived from i-Lats1/2鈭咶RC-TR mice. Following 4-hydroxytamoxifen (4-OHT) treatment, efficient (~83.7%) depletion of LATS1/2 and upregulation of YAP target genes were confirmed compared with control EtOH treatment (Fig.4e鈥揼 and Supplementary Fig.12a,b). In this condition, levels of pro-fibrotic markers such as vimentin and collagen I34 were increased, and contractility was potentiated as shown as enhanced assembly of F-actin (Fig.4h, i and Supplementary Fig.12c). To see whether this finding can be recapitulated in human, we cultured primary FRCs derived from human normal LNs. Primary cultured human FRCs transfected with adenovirus encoding constitutive active form of YAP (YAP5SA) or TAZ (TAZ4SA) enhanced the levels of 伪SMA, vimentin and collagen I compared with those transfected with control vector (Fig.4j鈥搇), indicating that phenotypes of YAP/TAZ hyperactivation in FRCs are instrinsic.YAP/TAZ regulate chemokine expression before LT尾R engagementTo examine the regulatory mechanism between LT尾R and YAP/TAZ, primary cultured FRCs derived from WT mice were stimulated with an agonistic LT尾R antibody (Supplementary Fig.13a,b). As expected, the stimulation of LT尾R increased NIK protein but reduced p100 protein in time and dose-dependent manners (Supplementary Fig.13b,c). Of note, it not only increased p52 protein level, but also promoted activity of LATS and increased pYAP/YAP ratio (Fig.5a). Importantly, nuclear鈥恈ytoplasmic fractionation and immunofluorescence analyses revealed that the LT尾R stimulation promoted nuclear to cytoplasmic shuttling of YAP/TAZ (Fig.5b and Supplementary Fig.13d).Thus,LT尾R signaling suppresses YAP/TAZ activity in FRCs.Fig. 5: YAP/TAZ regulate chemokine expression prior to LT尾R engagement.a Immunoblot analyses at indicated time points and comparison of normalized pYAP/YAP ratio at 240鈥塵in in cultured FRCs derived from WT mice after stimulation with LT尾R agonistic antibody (500鈥塶g/ml) for indicated time points. b Immunoblot analyses of indicated proteins in nuclear (LaminB) and cytoplasmic (GAPDH) fractions of cultured FRCs after treatment with or without LT尾R agonistic antibody. c Immunoprecipitation (IP) with anti-IgG or anti-YAP/TAZ (伪Y/T) antibody in primary cultured FRCs derived immunoblot with indicated antibodies. d Pull-down assay with streptavidin resin in HEK-293T cells after transfection with the streptavidin-binding peptide (SBP)-TAZ4SA, with or without plasmids encoding p52 or RelB and immunoblot analysis with indicated antibodies. e Pull-down assay with streptavidin resin in HEK-293T cells after transfection with the (SBP)-TAZ4SA, with or without plasmids encoding p52 (WT) or p52-Y293A mutants (YA) and immunoblot analysis with indicated antibodies. f Pull-down assay with streptavidin resin in HEK-293T cells after transfection with (SBP)-TAZ4SA or (SBP)-WW domain-deleted TAZ mutant (鈻?/span>WW) with or without plasmids encoding p52 or RelB and immunoblot analysis with indicated antibodies. g Diagram depicting the p52/RelB binding site within the mouse Ccl19 promoter and Ccl19 promoter-driven luciferase constructs containing p52/RelB binding site (WT) or the binding site deletion mutant (Mut). h Comparison of relative luciferase reporter activity using WT and Mut in HEK-293T cells. WT and Mut was co-transfected with or without p52 or p52 mutant (YA) and TAZ or TAZ mutant (鈻?/span>WW) in HEK-293T cells (n鈥?鈥?). P values by one-way ANOVA. i Representative images of in situ proximity ligation assay showing localizations of YAP or TAZ and p52 after treatment with or without LT尾R agonistic antibody in cultured FRCs. Nuclei are stained with DAPI. Scale bars, 50鈥壜祄. j ChIP experiments using IgG or anti-TAZ antibody were performed in MEFs infected with retrovirus encoding CTL or TAZ4SA with or without LT尾R agonistic antibody. Unless otherwise denoted, similar findings were observed in three independent experiments. Horizontal bars indicate mean鈥壜扁€塖D and P value versus 0鈥塵in or Control by two-tailed Student鈥檚 t-test. NS, not significant.Full size imageLT尾R intracellular signaling mainly takes a non-canonical NF-魏B pathway through p52/RelB to exert its cellular functions35,36. To examine whether YAP/TAZ physically interact with either p52 or RelB, we performed immunoprecipitation analysis using anti-YAP/TAZ antibody in primary cultured FRCs derived from WT mice. p52 was readily detected in the immuno-complexes pulled with anti-YAP/TAZ antibody but not in those from control IgG (Fig.5c). To ensure this finding, we transfected HEK293T cells with a gene encoding streptavidin-binding peptide-tagged constitutively active form of TAZ (SBP-TAZ4SA) and pulled all bound proteins using the streptavidin resin. We confirmed presence of both p52 and RelB in the resin, and stabilization of p52 protein through TAZ (Fig.5d and Supplementary Fig.13e).YAP/TAZ contain WW domains that bind to PPxY motif of the binding protein partners25,37. Although neither p52 nor RelB has PPxY motif, both p52 and RelB have highly conserved PPY motifs38,39, which could be possible binding targets for YAP/TAZ. Of note, point mutation of the PPY motif of p52 to PPA (p52-Y293A) completely abrogated the binding interaction with SBP-TAZ4SA (Fig.5e), whereas the RelB mutants (RelB-Y248A and RelB-Y341A) still showed interaction with SBP-TAZ4SA (Supplementary Fig.13f). Conversely, deleting the WW domain in SBP-TAZ4SA (TAZ4SA鈻?/span>WW) markedly attenuated its binding with p52, but again this did not affect the binding with RelB (Fig. 5f), indicating that TAZ independently binds to both RelB and p52. To elucidate how this molecular interaction regulates FRC differentiation, we transfected HEK293T cells retaining a Ccl19 promoter-driven luciferase with the gene encoding either TAZ4SA, TAZ4SA鈻?/span>WW, p52, or p52-Y293A (Fig. 5g). Intriguingly, while a single transfection of p52 enhanced the luciferase activity by 7.5-fold, co-transfection with TAZ4SA and p52 promoted the luciferase activity by 13.5-fold (Fig.5h). However, this increase in luciferase activity was not observed in cells that were transfected with either TAZ4SA鈻?/span>WW, p52-Y293A or both of these (Fig. 5h). We further validated our findings in primary cultured FRCs derived from i-Yap/Taz鈭咶RC mice, where p52-regulated transcripts such as Ccl19 and Ccl21 were markedly attenuated by 4-OHT treatment compared with EtOH (Supplementary Fig.13g,h). Conversely, in situ proximity ligation assay and chromatin immunoprecipitation (ChIP) analysis revealed that LT尾R stimulation promoted nuclear to cytoplasmic shuttling of YAP/TAZ-p52 complex and attenuated its binding affinity to promoter regions of Ccl19 (Fig.5i, j). Thus, proper interaction between YAP/TAZ and p52 seems to be required for the expression of chemokines such as Ccl19 before LT尾R engagement (Supplementary Fig.13i).YAP/TAZ drive specification of mesenchymal cells into FRCsThese observations led us to postulate that the binding of YAP/TAZ to p52 is required in myofibroblastic FRC precursors before LT尾R engagement. To verify this postulation, we depleted Yap/Taz in myofibroblastic FRC precursors (Ltbr鈭咶RC-YR) by generating Ltbr-Y/T鈭咶RC-YR mice, and we also generated i-Ltbr-Y/T鈭咶RC-YR mice by crossing Pdgfrb-CreERT2-YFP reporter mouse (i-WT鈭咶RC-YR) with Ltbrflox/flox (i-Ltbr鈭咶RC-YR) or Yapflox/flox/Tazflox/flox mouse (i-Y/T鈭咶RC-YR) (Fig.6a). Surprisingly, the skin-draining LNs including inguinal, axillary and brachial LNs of Ltbr-Y/T鈭咶RC-YR and i-LTbR-Y/T鈭咶RC-YR mice had perilipin+ adipocytes, which constitutes ~20鈥?5% of the LN density (Fig.6b, c and Supplementary Fig.14a).Fig. 6: Depletion of Yap/Taz transforms mesenchymal FRC precursors into adipocytes.a Diagram for analyses of indicated mice at 8-weeksold with or without tamoxifen delivery from 4-weeksold. b, c Representative images and comparisons of perilipin+ adipocytes within the inguinal LN (dashed line) in indicated mice (n鈥?鈥?). Scale bars, 400鈥壜祄. d Representative images of inguinal LN filled with adipocytes in Ltbr-Y/T鈭咶RC-YR mice (n鈥?鈥?). Right upper panel shows the magnified view of the region within the white dashed box and yellow arrowheads in the right lower panel indicate CCL19-YFP+perilipin+BODIPY+ adipocytes. Scale bars, 500鈥壜祄 (left panel); 100鈥壜祄 (right lower panel). e Representative images of perilipin+ adipocytes along the LYVE-1+ lymphatic vessels (dashed-boxes) in inguinal LN of i-Ltbr-Y/T鈭咶RC-YR mice (n鈥?鈥?). Scale bar, 400鈥壜祄. f, Diagram for primary culture of FRCs derived from i-Ltbr-Y/T鈭咶RC-YR mice for 4 days and treatment with EtOH or 4-OHT for their analyses at 2 days after the treatment. g Comparisons of indicated mRNA expression normalized to Gapdh in primary cultured FRCs after treatment with EtOH or 4-OHT for 2 days (n鈥?鈥?). h Diagram for adipogenic culture of mesenchymal stem cells (C3H/10T1/2) infected with an adenovirus to induce overexpression of active TAZ (TAZ4SA) for their analyses at 8 days after the infection. i Immunoblot analyses of indicated proteins in mesenchymal stem cells (C3H/10T1/2) infected with an adenovirus to induce overexpression of active TAZ (TAZ4SA) or control. j Representative images of Oil Red O staining in mesenchymal stem cells (C3H/10T1/2) induced with adipogenic cocktail after infected with an adenovirus to induce overexpression of active TAZ (TAZ4SA). k Comparisons of indicated mRNA expression normalized to Gapdh in mesenchymal stem cells (C3H/10T1/2) infected with an adenovirus to induce overexpression of active TAZ (TAZ4SA) for their analyses at 2 days after the infection (n鈥?鈥?). l Schematic images proposing the importance of coordination of YAP/TAZ activity and LT尾R coupling in FRCs during LN growth and maintenance. Unless otherwise denoted, horizontal bars indicate mean鈥壜扁€塖D and P values versus non-Y/T鈭咶RC-YR or non-i-Ltbr鈭咶RC-YR or EtOH or Control by two鈥恡ailed Mann鈥怶hitney U test.Full size imageThese aberrant adipocytes were mainly located along the infiltrating lymphatic vessels within the skin-draining LNs, while no adipocytes were observed within the mesenteric LNs of Ltbr-Y/T鈭咶RC-YR and i-Ltbr-Y/T鈭咶RC-YR mice (Supplementary Fig.14a). Of note, all the perilipin+ or BODIPY+ adipocytes were YFP+ in LNs of Ltbr-Y/T鈭咶RC-YR and i-Ltbr-Y/T鈭咶RC-YR mice, indicating that they originated from YFP+ FRC precursors (Fig.6d, e). Further analysis with primary cultured FRCs derived from Ltbr-Y/T螖FRC-YR mice showed upregulation of adipogenic genes such as Pparg and Cebpa40 without ectopic fat accumulation (Fig.6f, g, and Supplementary Fig.14b), suggesting enhanced adipogenic activity of FRCs rather than metabolic dysregulation in Ltbr-Y/T螖FRC-YR mice.It has been proposed that both adipocytes and LN stromal cells are mesenchymal origin and developmentally related12. We therefore examined whether Hippo signaling determines the fate specification of mesenchymal stem cells by utilizing C3H10T1/2 cells (Fig.6h). Adipogenic culture of C3H10T1/2 cells led to their differentiation into adipocyte-lineage cells, but TAZ4SA overexpressing C3H10T1/2 cells significantly abrogated adipogenesis while they expressed enhanced FRC commitment markers19,41 (Fig.6i鈥搆). These data indicate that YAP/TAZ drive specification and differentiation of mesenchymal cells into FRCs, while they suppress those into other cell types including adipocytes.DiscussionFRCs discard some of their myofibroblastic characters as they differentiate and complete maturation, while they gain chemokine secretory functions which are essential for organizing adaptive immunity12,20. Here, we show that the canonical Hippo pathway, together with LT尾R-non-canonical NF-魏B signaling, critically regulates differentiation and maintenance of LN FRCs (Fig.6l). Deletion of YAP/TAZ in FRCs during development impairs their growth and differentiation, compromising the structural organization of LNs. Hyperactivation of YAP/TAZ in FRCs during the developmental period severely impairs differentiation and maturation of FRCs, leaving non-functional and fibrotic LNs constituted with immature FRCs. Even when YAP/TAZ are hyperactivated in adult mature FRCs, enhanced myofibroblastic characters of FRCs and severe LN fibrosis are similarly observed. These alterations ultimately lead to substantial distortion of LN microarchitecture with impairs adaptive immune responses. Although depletion or hyperactivation of YAP/TAZ leads to similar phenotypes, outcomes of YAP/TAZ depletion are due to loss of FRC pool, while those of YAP/TAZ hyperactivation are owed to maturation defects of FRCs.These results show that proper modulation of FRCs by canonical Hippo signaling is critical for formation and maintenance of LNs during development and in adult. A previous study proposes that mesenchymal stem cells are differentiated into mature FRCs by suppressing their adipogenicity through promotion of LT尾R-p52/RelB signaling12. TAZ binding to PPAR纬 via its WW domain is also demonstrated to inhibit adipogenesis42. However, the association between LT尾R signaling and YAP/TAZ has not been previously addressed. Our biochemical analyses demonstrate that YAP/TAZ and p52 form a complex to regulate the expression of chemokines such as CCL19 in FRC precursors. Then, upon LT尾R activation, YAP/TAZ become phosphorylated and translocated to the cytoplasm, leading to the maturation of FRC precursors. While previous studies demonstrate that LT尾R signaling is critical for FRC differentiation, neither ablation of LT尾R nor its ligands in FRC precursors is critical for maintaining the FRC lineage19,20,21. Here, we show that depletion of YAP/TAZ in LT尾R-ablated FRC precursors induces the transition of the FRC lineage into adipocytes. Collectively, these results indicate that LT尾R signaling and YAP/TAZ play both overlapping and independent roles for cell commitment and maintenance of FRCs.Still, questions remain in linking Yap/Taz modulation in FRCs with its final outcomes. In case of Yap/Taz depletion during development, it results in decreased cellularity of FRCs, which lead to shortage in conduit coverage. Considering that this study agrees with a previous study in that FRC coverage of the LN conduit is not critical for the maintenance of conduit integrity and function17, the purpose of FRC coverage around conduits needs further investigation. In contrast, Yap/Taz activation during development lead to maturation defects, and as a consequence impairs LN microarchitecture, which itself could not fully explain the lethal phenotypes. Therefore, the effects of Yap/Taz modulation in other FRC subsets43,44 in lymphoid tissues such as spleen and thymus remains to be studied. In addition, exploiting a more specific inducible Cre driver would provide better understanding when comparing the consequences of constitutional or conditional YAP/TAZ manipulation in FRCs. In conclusion, our study clearly demonstrates that the Hippo pathway plays pivotal roles in maturation and maintenance of FRCs of LNs.MethodsMiceSpecific pathogen-free (SPF) C57BL/6J mice (#000664), Rosa26-tdTomato mice (#007914), Rosa26-eYFP (#007903), Actb-DsRed (#006051), Actb-GFP (#003291), Lgr5-Cre (#008875), and OT-II (#004194) mice were purchased from the Jackson Laboratory. Lats1flox/flox鈥?sup>29, Lats2flox/flox鈥?/sup>30, Yapflox/flox鈥?/sup>27, Tazflox/flox鈥?/sup>28, Ltbrflox/flox鈥?sup>20, Ccl19-Cre20, and Pdgfrb-Cre-ERT2鈥?sup>45 mice were transferred, established, and bred in SPF animal facilities at KAIST. All mice were maintained in the C57BL/6 background and fed with free access to a standard diet (PMI LabDiet) and water. In order to induce Cre activity in the Cre-ERT2 mice, 2鈥塵g of tamoxifen (Sigma-Aldrich) was dissolved in corn oil (Sigma-Aldrich) and intra-peritoneally (i.p.) injected at indicated time points. All mice were anesthetized with i.p. injection of a combination of anesthetics (80鈥塵g/kg ketamine and 12鈥塵g/kg of xylazine) before being euthanatized. We complied with all ethical regulations for animal testing and research and performed all animal experiments and euthanasia under the approval from the Institute Animal Care and Use Committee (No. KA2016-12) of Korea Advanced Institute of Science and Technology (KAIST). Mouse model nomenclatures are included in Supplementary Table1.Histological analysesGross images of LNs were acquired using AxioZoom V16 stereo zoom microscope (Carl Zeiss). For LN weight measurement, bilateral inguinal LNs were pooled and weighed using an analytical balance (Mettler Toledo). For immunofluorescence staining of LNs, harvested samples were fixed in 1% paraformaldehyde (PFA) in PBS overnight at 4鈥壜癈 and dehydrated in 20% sucrose in PBS overnight at 4鈥壜癈. Samples were embedded in tissue freezing medium (Leica) and frozen blocks were cut into 20-渭m-thick sections. Samples were blocked with 5% goat (or donkey) serum in 0.3% Triton X-100 in PBS (PBST) and incubated for overnight at 4鈥壜癈 with primary antibodies (diluted at a ratio of 1:200 in blocking solution). After several washes with PBS, samples were incubated for 2鈥塰 at RT with secondary antibodies (diluted at a ratio of 1:1000 in blocking solution). After washing several times with PBS, samples were mounted with fluorescent mounting medium (DAKO) and images were acquired using LSM780 or LSM880 confocal microscope (Carl Zeiss).To examine YAP and TAZ distribution in human LNs, several cervical LNs around thyroid papillary carcinoma were collected from the patients undergoing thyroidectomy with written informed consent according to the protocol approved by the institutional review board of Pusan National University (H-1610-002-003) and Samsung Medical Center (2018-06-061).For whole-mount staining of embryo, pregnant mice were anesthetized at the indicated day of embryonic development after vaginal plug, and the uteri were removed. Embryos were dissected and the yolk sacs were used for verification of genotype. Inguinal fat pads and LN anlagen were isolated together with the covering skin and processed for staining. Samples were fixed in 2% PFA for 2鈥塰 at 4鈥壜癈 and washed several times with PBS for the staining.For staining of primary cultured FRCs, cells were plated in 8-well Nunc Lab-Tek II chamber slides (Sigma-Aldrich) and fixed with 4% PFA for 8鈥塵in at RT. After several washes with PBS, samples were blocked with 5% goat (or donkey) serum in PBST for 30鈥塵in at RT. Cells were incubated with primary antibodies (diluted at a ratio of 1:200 in blocking solution) for overnight at 4鈥壜癈. The following primary and secondary antibodies were used in the immunostaining: anti-YAP (rabbit monoclonal, D8H1X, Cell Signaling), anti-TAZ (rabbit polyclonal, HPA007415, Sigma-Aldrich), anti-PDPN (syrian hamster monoclonal, 127402, Biolegend), anti-CCL19 (goat polyclonal, PA5-47958, Thermo Fisher), anti-CCL21 (goat polyclonal, AF457, R D), anti-ER-TR7 (rat monoclonal, sc-73355, Santa Cruz), anti-CD3e (hamster monoclonal, 145-2C11, BD), anti-B220 (rat monoclonal, RA3-6B2, BD), anti-CD31 (hamster monoclonal, 2H8, Millipore), anti-PDGFR尾 (rat monoclonal, APB5, eBioscience), anti-collagen IV (rabbit polyclonal, ab6586, Abcam), anti-Ki-67 (rabbit monoclonal, SP6, Abcam), anti-LYVE-1 (rabbit polyclonal, 11-034, Angiobio), anti-caspase-3 (rabbit polyclonal, 9661, Cell Signaling), anti-vimentin (chicken polyclonal, AB5733, Millipore), anti-collagen I (rabbit polyclonal, ab34710, Abcam), anti-LT尾R (rabbit polyclonal, ab70063, Abcam), anti-CD4 (rat monoclonal, GK1.5, BD), anti-CD11b (rat monoclonal, M1/70, BD), anti-CD11c (hamster monoclonal, N418, Bio-Rad), anti-perilipin (guinea pig polyclonal, 20R-PP004, Fitzgerald), anti-PNAd (rat monoclonal, MECA-79, BD), anti-ICAM1 (rat monoclonal, YN1/1.7.4, Abcam), anti-Prox1 (rabbit polyclonal, 102-PA32, ReliaTech), and FITC- or Cy3-conjugated anti-伪SMA (mouse monoclonal, 1A4, Sigma-Aldrich). FITC-, Cy3- or Cy5-conjugated secondary antibodies were purchased from Jackson ImmunoResearch. Lipids were stained with BODIPY (Invitrogen) and nuclei were stained with DAPI (Invitrogen).Immunological analysisFor adoptive transfer of labeled cells, mononuclear cells were isolated from the spleen and skin-draining LNs from Actb-DsRed and Actb-GFP mice. B cells from Actb-DsRed mice and T cells from Actb-GFP mice were further sorted using Dynabeads Untouched Mouse T Cells Kit and Pan B cell Kit (Thermo Fisher). Mixture of 1鈥壝椻€?07 T cells and 1鈥壝椻€?07 B cells were transferred intravenously into the mouse. LNs of the recipient mice were isolated and analyzed at 24鈥塰 after the adoptive transfer.For influenza viral infection, mice were immunized by intranasal administration of 1鈥壝椻€?03鈥塸fu of A/PR/8 [A/Peuerto Rico/8/34 (H1N1)]. Then the mice were monitored for weight loss and survival every day for the next 14 days. For detection of influenza-specific antibodies, ELISA plates (Falcon) coated with 20鈥壩糶/ml of formalin-inactivated A/PR/8 virus in PBS were incubated overnight at 4鈥壜癈. Blocking with 1% BSA in PBS was performed at 37鈥壜癈 for 1鈥塰. Serial two-fold dilutions of samples (starting from 1:64 of serum) were applied to plates and incubated for 4鈥塰 at 37鈥壜癈. Horseradish peroxidase-conjugated goat anti-mouse IgG antibodies (Southern Biotechnology Associates) were added to each well and incubated for 2鈥塰 at 37鈥壜癈. For color development, TMB-H2O2 solution (Moss) was added as chromogen substrate to each well and incubated for 15鈥塵in at RT. Once stopping solution had been added (0.5鈥塏 HCl), absorbance was measured at 450鈥塶m using an ELISA reader (Molecular Devices). Endpoint titers of A/PR/8 virus-specific antibodies were expressed as reciprocal log2 titers of the highest dilution that showed a level of 0.1 absorbance over background.For intracellular cytokine staining, BD Cytofix/Cytoperm Plus (BD Pharmigen) was used according to the manufacturer鈥檚 instructions. For the measurement of IFN纬-producing CD8+ T cells, mononuclear cells were obtained from mediastinal LNs and were incubated with 5鈥塶g/ml of PMA and with 500鈥塶g/ml of Ionomycin in the presence of GolgiPlug (BD Pharmingen) for 4鈥塰. The cells were stained with anti-CD8 antibody (rat monoclonal, 53-6.7, BD) and anti-IFN-g antibody (rat monoclonal, XMG1.2, BD), and analyzed with FACS Canto II (BD Biosciences).For systemic depletion of T cells, anti-mouse CD33 monoclonal Ab (1鈥塵g/kg/day for 5 days; Clone 145-2C11, ATCC) was intravenously administered into WT mice. As a control, the same amount of hamster anti-mouse IgG isotype Ab (R D Systems) was administered in the same manner.For analysis of T cell proliferation in vivo, OVA-specific T cells were isolated from OT-II mice, labeled with 10鈥壩糓 CFSE, and then adoptively transferred intravenously into the mice. Twenty-four hours after the transfer, 100鈥壩糶 of OVA was injected via footpad. Three days after the injection, mice were euthanatized and mononuclear cells were isolated from popliteal LNs. Activation and CFSE dilution of adaptive transferred OT-II CD4+ T cells were analyzed by flow cytometry.Electron microscopyLNs were fixed in 2.5% glutaraldehyde in 0.1鈥塎 phosphate buffer (pH 7.4) overnight at 4鈥壜癈 and post-fixed with 1% osmium tetroxide for 2鈥塰 at RT. Samples were dehydrated with series of increasing ethanol concentrations followed by resin embedding. 70-nm-thick ultrathin sections were obtained with an ultramicrotome (UltraCut-UCT, Leica), which were then collected on copper grids. After staining with 2% uranyl acetate and lead citrate, samples were examined by transmission electron microscope (Tecnai G2 Spirit Twin, FEI) at 120鈥塳V.LN conduit analysisConduit staining was achieved by injecting 10鈥壜礚 of saturated FITC (Sigma-Aldrich) in HBSS (0.1鈥塵g/mL) into the footpad, and the draining popliteal LN was collected at 10鈥塵 after the injection. Samples were fixed in 1% PFA for 1鈥塰 immediately after the harvest and processed for whole-mount and imaging.Morphometric analysesMorphometric analyses of the LNs were performed with images taken from mid-sectioned LNs by photographic analysis using ImageJ software (http://rsb.info.nih.gov/ij) or ZEN 2012 software (Carl Zeiss). Numbers of Ki-67+ proliferating or caspase-3+ apoptotic FRCs, T cells, and B cells were measured from random four or five 0.045鈥塵m2 fields of LNs and averaged. Nuclear YAP was counted by calculating the proportion of PDGFR尾+ FRCs with nuclear YAP from random four or five 0.045鈥塵m2 fields of LNs and averaged. The relative expressions of indicated markers were calculated as PDPN+, 伪SMA+, PDGFR尾+, collagen IV+, CCL19+ or CCL21+ area divided by Tomato+, YFP+ or PDGFR尾+ area. For density measurements, PDPN+, 伪SMA+, PDGFR尾+, collagen IV+ or ER-TR7+ area was measured in random four or five 0.045鈥塵m2 area and presented as a percentage of the total measured area. FRC surface area was assessed by employing a 3D reconstruction analysis using Imaris (Bitplane)46. The surface area of Tomato+ FRCs was calculated using the single-cell morphometric 鈥榮urface鈥?module in Imaris after the 3D reconstruction of a given image and volume filter was used to reduce background noise. Random four or five 0.045鈥塵m2 fields of LNs were analyzed and individual FRC was distinguished as a separate 3D surface object by using the 鈥榗utting鈥?tool and DAPI staining was utilized to identify the cell nuclei belonging to each FRC. Percentage of naked conduit and fibril irregularity were measured in up to eight random area of immunofluorescence or TEM images and averaged. Fibril irregularity was assessed by employing a previously described method47. Longitudinal section of TEM image was analyzed for non-condensed fibril bundles and unpacked collated fibers with discontinuous, multi-directional pattern were defined as irregular fibrils.Flow cytometry and cell sortingSkin-draining LNs (axillary, brachial, cervical, and inguinal LNs) were harvested and cut into small pieces for digestion48. Briefly, LNs were digested in 2鈥塵l of enzyme buffer containing 2鈥塵g/ml collagenase type II (Worthington Biochem), 0.1鈥塵g/ml DNase (Roche), and 1鈥塵g/ml dispase (Gibco) at 37鈥壜癈 for 30鈥塵in. Tissues were gently agitated and pipetted 1鈥? times during digestion to disrupt any cell clumps. When LNs were completely digested, cell suspension was filtered through 40鈥壩糾 nylon cell strainer and washed. Cells were incubated for 20鈥塵in with anti-CD45 Microbeads (Miltenyi). To enrich the stromal cell fraction, hematopoietic cells were depleted using AutoMACS (Miltenyi), according to the manufacturer鈥檚 instructions. The cells isolated from skin-draining LNs, spleen, and bone marrow were filtered through a 40鈥壩糾 nylon mesh to remove cell clumps. After RBC lysis by suspension in ACK lysis buffer for 5鈥塵in at RT, the cells were incubated for 30鈥塵in with antibodies (diluted at a ratio of 1:200) in FACS buffer (5% bovine serum in PBS). After several washes, cells were analyzed by FACS Canto II (BD Biosciences) and the acquired data were further evaluated by using FlowJo software (Treestar). Cell sorting was performed with FACS Aria II (Beckton Dickinson). Dead cells were excluded using DAPI staining (Sigma-Aldrich). The following antibodies were used other than the antibodies described above: anti-CD45 (rat monoclonal, 30-F11, eBioscience), anti-TER-119 (rat monoclonal, TER-119, eBioscience), anti-PDPN (syrian hamster monoclonal, 8.1.1, Biolegend), anti-CD31 (rat monoclonal, MEC 13.3, BD), and anti-CD19 (rat monoclonal, 6D5, Biolegend).For cell cycle analysis, mice were injected with 1鈥塵g of BrdU solution. Skin鈥恉raining LNs were harvested at 16鈥塰 after the injection of BrdU and isolated cells were processed with the APC BrdU flow kit (BD Biosciences) according to the manufacturer鈥檚 protocols. The cell cycle profiles were analyzed with a FACS Canto II (BD Biosciences) and assessed with FlowJo (Treestar).ATAC sequencingApproximately 20,000 tdTomato+ Lgr5+ stem cells in dermis were FACS-purified and used for ATAC-seq. The ATAC-seq libraries were prepared by employing a previously described method49. 2鈥壝椻€?01 paired-end sequencing was performed on Illumina HiSeq-2500. We randomly sampled 2.5鈥塎 reads from each sample using samtools view and pooled them into one file so that each sample is equally represented. Peaks were called on the pooled file as discussed in the previous paragraph. We then determined the number of samples overlapping with each master peak using peaks called on individual samples.Quantitative RT-PCR and RNA sequencingFor quantitative RT-PCR, total RNA was extracted from sorted cells by using Trizol RNA extraction kit (Invitrogen) according to the manufacturer鈥檚 instructions. Total RNA was reverse transcribed into cDNA using GoScript Reverse Transcription Kit (Promega). Then, quantitative real-time PCR was performed using FastStart SYBR Green Master mix (Roche) and Bio-rad S1000 Thermocycler. GAPDH was used as a reference gene and the results were presented as relative expressions to control. List of primer sequences are described in Supplementary Table2.For RNA sequencing, the construction of library was performed using QuantSeq 3鈥?mRNA-Seq Kit (Lexogen GmbH, Austria) according to the manufacturer鈥檚 instructions. In brief, each 500鈥塶g total RNA were prepared and an oligo-dT primer containing an Illumina-compatible sequence at its 5鈥?end was hybridized to the RNA and reverse transcription was performed. After degradation of the RNA template, second strand synthesis was initiated by a random primer containing an Illumina-compatible linker sequence at its 5鈥?end. The double-stranded library was purified by using magnetic beads to remove all reaction components. The library was amplified to add the complete adapter sequences required for cluster generation. The finished library was purified from PCR components. High-throughput sequencing was performed as single-end 75 sequencing using NextSeq 500 (Illumina). Mapping of RNA-Seq reads were performed using Bowtie2 version 2.1.0, which was used to bring together transcripts, assess their exuberances, and identify DEGs or isoforms using cufflinks. The RT (Read Count) data were processed based on Quantile normalization method using the Genowiz version 4.0.5.6 (Ocimum Biosolutions). The Ingenuity Pathway Analysis tool (QIAGEN) was used to interpret data in the context of canonical pathways, biological processes and networks. Both up- and downregulated identifiers were defined as value parameters for the analysis. Significance of the canonical pathways and biological function and networks were tested by the Fisher Exact test p-value, and determined for their activation or inhibition upon activation z-score. For GSEA, gene sets from the Molecular Signatures Database (MSigDB) 5.2 were used for the analysis of the Hippo pathway-responsive transcriptomes. Cluster analysis and heatmap construction were performed using Multiple Experiment Viewer (MeV) from The Institute of Genomic Research (TIGR) and Cluster and TreeView from the Eisen laboratory. Original data are available in the National Center for Biotechnology Information鈥檚 Gene Expression Omnibus (accession number GSE89742). For presentation of RNA-sequencing data into bar graphs, normalized raw counts (log2) were converted by using the following formula: 2(x: normalized raw count). Next, given values were divided by the average value of WT and presented as relative ratio.Primary culture of FRCsSkin-draining lymph nodes of indicated mice were harvested, cut into small pieces, and digested48. LNs were digested in 2鈥塵l of enzyme buffer containing mixture of 2鈥塵g/ml collagenase type II (Worthington Biochem), 0.1鈥塵g/ml DNase (Roche), and 1鈥塵g/ml dispase (Gibco) at 37鈥壜癈 for 30鈥塵in. Tissues were gently agitated and pipetted several times during digestion to disrupt any cell clumps. When LNs were completely digested, cell suspension was filtered through 40鈥壩糾 nylon cell strainer and washed. Cells were incubated for 20鈥塵in with anti-CD45 Microbeads (Miltenyi). To enrich the stromal cell fraction, hematopoietic cells were depleted using AutoMACS (Miltenyi), according to the manufacturer鈥檚 instructions. In order to induce FRC differentiation, cells were incubated with DMEM containing 10% fetal bovine serum and 500鈥塶g/ml of LT尾R agonistic antibody (Abcam, Cat. ab65089) for 5 days. For induction of cre activity in primary cultured FRCs derived from Lats1/2i鈭咶RC or Yap/Taz鈭咶RC mice, cells were treated with 5鈥壜礛 4-hydroxy-tamoxifen (4-OHT) in 100% ethanol (EtOH) or 100% EtOH alone for 2 days as a control. For primary culture of human FRCs, cervical LN specimens were acquired for diagnostic purposes from presently healthy adults with written informed consent according to the protocol approved by the institutional review board of Samsung Medical Center (2018-06-061). LNs were cut into small pieces immediately after the harvest and digested50. Following the digestion, cells were covered with media on a tissue culture plate and grown for 4 days to permit fibroblasts to emerge. Culture-expanded monolayer of FRCs under passage 3 that were validated as CD45鈭?/sup>, CD31鈭?/sup>, and PDPN+ were used for the experiments.For adenoviral infection, human TAZ4SA and YAP5SA cDNAs51 were cloned into the pAdtrack-CMV-GFP vector52. Cloning vectors were then recombined with the pAdEasy-1 vector in BJ5183-AD-1 electroporation鈥揷ompetent cells (Agilent Technologies). The recombinant DNA was linearized with PacI and introduced into 293AD cells by transfection with Lipofectamine LTX and PLUS reagent (Invitrogen). After verification of GFP expression, cell pellets were centrifuged and resuspended with 10% glycerol in PBS and lysed with 4 freeze-thaw cycles. Adenoviruses were purified by ultracentrifugation at 46,000鈥壝椻€?i>g for 2鈥塰 at 4鈥壜癈 within a discontinuous gradient from 2.2 to 4.0鈥塎 CsCl (Amresco) in 10鈥塵M HEPES (Sigma-Aldrich). The adenovirus-containing layer was removed with a syringe needle, and the viruses were washed twice in a solution containing 10鈥塵M Tris-HCl (pH 8.0) and 2鈥塵M MgCl2 using an Amicon Ultra Centrifugal Filter (Sigma-Aldrich). Virus titration was performed by counting exposed 293AD or target cells positive for GFP with a fluorescence microscope.ImmunoblottingFor immunoblot analysis, cells were lysed on ice in RIPA lysis buffer supplemented with protease and phosphatase inhibitors (Roche). Cell lysates were centrifuged for 10鈥塵in at 4鈥壜癈, 13,000鈥塺pm. Protein concentrations of the supernatants were quantitated using the detergent-insensitive Pierce BCA protein assay kit (Thermo Fischer, 23227). Laemmli鈥檚 buffer was added to total protein lysates and samples were denatured at 95鈥壜癈 for 5鈥塵in. Aliquots of each protein lysate (10鈥?0鈥壩糶) were subjected to SDS polyacrylamide gel electrophoresis. After electrophoresis, proteins were transferred to nitrocellulose membranes and blocked for 30鈥塵in with 5% skim milk in TBST (0.1% Tween 20 in TBS). For phosphorylated protein detection, membranes were blocked with 2% BSA in TBS. For detection of protein stability, protein synthesis was blocked by treatment of 50鈥壩糶/ml cycloheximide (CHX) for indicated time. Primary antibodies (diluted at a ratio of 1:1000 in blocking solution) were incubated overnight at 4鈥壜癈. After washes, membranes were incubated with anti-rabbit (CST, #7074) or anti-mouse (CST, #7076) secondary peroxidase coupled antibody (diluted at a ratio of 1:5000 in TBST) for 1鈥塰 at RT. Target proteins were detected using ECL western blot detection solution (Millipore, WBKLS0500). The following antibodies were used for immunoblotting: anti-YAP (rabbit monoclonal, D8H1X, Cell Signaling), anti-phospho-YAP (rabbit polyclonal, 4911, Cell Signaling), anti-YAP/TAZ (rabbit monoclonal, D24E4, Cell Signaling), anti-TAZ (rabbit monoclonal, V386, Cell Signaling), anti-LATS1 (rabbit monoclonal, C66B5, Cell Signaling), anti-phospho-LATS1/2 (rabbit monoclonal, D57D3, Cell Signaling), anti-CTGF (rabbit polyclonal, ab6992, Abcam), anti-RelB (rabbit monoclonal, D7D7W, Cell Signaling), anti-CTGF (rabbit polyclonal, ab6992, Abcam), anti-p100/p52 (rabbit polyclonal, 4882, Cell Signaling), anti-p100/p52 (rabbit monoclonal, sc-7386, Santa Cruz), anti-NIK (rabbit polyclonal, 4994, Cell Signaling), anti-LaminB (rabbit monoclonal, D4Q4Z, Cell Signaling), anti-GAPDH (rabbit monoclonal, D16H11, Cell Signaling), and anti-尾 actin (rabbit monoclonal, AC-74, Sigma-Aldrich).For nuclear鈥恈ytoplasmic fractionation of cells, harvested cells were resuspended in hypotonic lysis buffer (10鈥塵M HEPES [pH 7.8], 10鈥塵M KCl, 1.5鈥塵M MgCl2, 0.5鈥塵M DTT, and protease inhibitors). To break plasma membrane, the resuspended cells were mixed with 0.3% NP-40 by vortexing for 5鈥塻, and the cytoplasmic fraction was obtained from the supernatant after centrifugation. After several washes with PBS, the pellet was boiled in Laemmli sample buffer and used as the nuclear fraction. The uncropped and unprocessed scans with marker positions of all blots were included in the Source Data file.Pull-down assay and co-immunoprecipitationThe N-terminal Flag-SBP tagged human TAZ4SA cDNAs were cloned into the pcDNA3.1 vector (Thermo Fischer, Cat. V790-20). Flag-p52 and Flag-RelB were purchased from Addgene (plasmid # 20019, # 20017). Each mutant construct (TAZ4SA鈻?/span>W, p52 YA and RelB YA) was generated by overlap extension PCR. HEK293T cells were cultured in 6鈥恮ell plate and were co-transfected with indicated constructs using polyethylenimine Max (Polyscience, Cat. 24765-1). Two days after transfection, the cells were harvested, lysed with NETN buffer (20鈥塵M Tris-HCl (pH 7.4), 100鈥塵M NaCl, 1鈥塵M EDTA, 0.5% Nonidet P-40, and protease inhibitors). Cell extracts (1鈥塵g) were incubated with Streptavidin Agarose (Pierce, Cat. 20359) for 2鈥塰 at 4鈥壜癈, and then the beads were washed three times with lysis buffer and boiled with Laemmli sample buffer. For co-immunoprecipitation, harvested FRCs were lysed with NETN buffer. Cell extracts (0.5鈥塵g) were incubated overnight at 4鈥壜癈 with 0.5鈥壩糶 of anti-IgG (rabbit polyclonal, H-270, Santa Cruz) or anti-YAP/TAZ (rabbit monoclonal, D24E4, Cell Signaling) antibody. The extracts were incubated with 20鈥壩糽 of protein A/G agarose beads (Pierce) for 1鈥塰, and then the beads were washed three times with lysis buffer (Triton X鈥?00 reduced to 0.1%) and boiled with Laemmli sample buffer. For detection of p52, the TrueBlot anti-Rabbit IgG HRP (ROCKLAND, Cat. 18-8816-31) was used as secondary antibody for reducing the heavy chain interference.Luciferase assayThe indicated portion of the Ccl19 genomic locus including p52 binding site was cloned into the pGL3-Basic vector (Promega, Cat. E1751). Mutant construct was generated with deletion of p52 binding site by overlap extension PCR. HEK293T cells were cultured in 24鈥恮ell plate and were co-transfected with a 100鈥塶g HA鈥怲AZ4SA (WT or 鈻?/span>W), 100鈥塶g Flag鈥恜52 (WT or YA), 200鈥塶g CCL19 luciferase (WT or Mut) or 20鈥塶g CMV鈥怰enilla per well by using polyethylenimine Max. Twenty-four hours later, cells were harvested, lysed, and assayed with the Dual Luciferase Reporter Assay System (Promega, Cat. E1960). List of primer sequences for luciferase assay are described in Supplementary Table 2.ChIP-qPCR analysisMEF cells were fixed with 1% formaldehyde for 10鈥塵in and then neutralized with 125鈥塵M glycine for 5鈥塵in at RT. The cells were washed with PBS and then lysed with ChIP dilution buffer [50鈥塵M HEPES (pH 7.5), 155鈥塵M NaCl, 1% Triton X-100, 0.1% sodium deoxycholate, 123鈥塵M EDTA] containing 1% SDS. The DNA in the cell lysates was fragmented by sonication using a bioruptor sonicator. The cell lysates were centrifuged at 13,000鈥塺pm for 15鈥塵in at 4鈥壜癈, and the resulting supernatants were further diluted with ChIP dilution buffer. They were then incubated overnight at 4鈥壜癈 with either TAZ antibody (Sigma, HPA007415) or IgG (Santa Cruz Biotechnology). The next day, protein A/G beads (Gendepot) were added and the samples were incubated for an additional 3鈥塰 at 4鈥壜癈. The beads were then isolated with centrifugation, washed with ChIP wash buffer [10鈥塵M Tris-HCl (pH 8.0), 250鈥塵M LiCl, 0.5% Nonidet P-40, 0.5% sodium deoxycholate, 1鈥塵M EDTA], and suspended in SDS lysis buffer [50鈥塵M Tris-HCl (pH 8.0), 10鈥塵M EDTA, 1% SDS] for overnight incubation at 65鈥壜癈. The beads were then removed, and the remaining material was incubated for 2鈥塰 at 55鈥壜癈 with proteinase K (20鈥塵g/ml) and glycogen (20鈥塵g/ml). After a final 1鈥塰 incubation with RNaseA at 37鈥壜癈, the DNA was purified using standard procedures and analyzed for 鈥淕GGRNNYYCC鈥?motif. List of primer sequences for ChIP-qPCR are described in Supplementary Table2.In situ proximity ligation assayCells cultured on confocal dishes were fixed with 4% paraformaldehyde for 20鈥塵in at RT, permeabilized and incubated with primary antibodies at 4鈥壜癈. For in situ proximity ligation assay, protein鈥損rotein interactions between YAP or TAZ and p52 were detected with secondary proximity probes (Anti-Rabbit Plus and Anti-Mouse Minus) according to the Duolink in situ Fluorescence User Guide (Sigma-Aldrich).Adipogenic inductionIn order to induce adipogenic differentiation, confluent cells were incubated with adipogenic differentiation medium [DMEM containing 10% fetal bovine serum, 5鈥壩糶/ml insulin (Sigma-Aldrich), 0.5鈥塵M 3-isobutyl-1-methylxanthine (IBMX, Sigma-Aldrich) and 1鈥壩糓 dexamethasone (Sigma-Aldrich)]. After 3 days, adipogenic differentiation medium was replaced with maintenance medium (DMEM containing 10% fetal bovine serum and 5鈥壩糶/ml insulin). For Oil Red O staining, cells were fixed with 4% paraformaldehyde for 40鈥塵in at RT. After being washed with PBS and 60% isopropanol, cells were incubated with filtered Oil Red O working solution for 50鈥塵in at RT. After staining, several washes with PBS and 60% isopropanol were performed to reduce non-specific staining. Images of stained cells were captured by a microscope equipped with a CCD camera (Carl Zeiss).Statistical analysesAll values are presented as mean鈥壜扁€塻tandard deviation (SD). For continuous data, statistical significance was determined by the Mann鈥揥hitney U test or Student鈥檚 t-test between two groups and one-way ANOVA for multiple-group comparison. The survival curve was evaluated using the Kaplan鈥揗eier method, and the statistical difference was analyzed using the log-rank test. Statistical analysis was performed with GraphPad Prism. Statistical significance was set to P鈥?lt;鈥?.05.Reporting summaryFurther information on research design is available in theNature Research Reporting Summary linked to this article. The RNA-sequencing and ATAC-sequencing data generated with this study have been deposited in Gene Expression Omnibus under the accession number GSE89742. The source data underlying all Figs. and Supplementary Figs. are provided as a Source Data file. A Reporting summary for this article is available as a聽Supplementary Information file. All other data that support the findings of this study are available from the corresponding author upon reasonable request. References1.Fletcher, A. L., Acton, S. E. Knoblich, K. Lymph node fibroblastic reticular cells in health and disease. Nat. Rev. Immunol. 15, 350鈥?61 (2015).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 2.Petrova, T. V. Koh, G. Y. Organ-specific lymphatic vasculature: from development to pathophysiology. J. Exp. Med. 215, 35鈥?9 (2018).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 3.Chang, J. E. Turley, S. J. Stromal infrastructure of the lymph node and coordination of immunity. Trends Immunol. 36, 30鈥?9 (2015).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 4.Bajenoff, M., Glaichenhaus, N. Germain, R. N. Fibroblastic reticular cells guide T lymphocyte entry into and migration within the splenic T cell zone. J. Immunol. 181, 3947鈥?954 (2008).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 5.Siegert, S. Luther, S. A. Positive and negative regulation of T cell responses by fibroblastic reticular cells within paracortical regions of lymph nodes. Front Immunol. 3, 285 (2012).PubMed聽 PubMed Central聽 Article聽Google Scholar聽 6.Cremasco, V. et al. B cell homeostasis and follicle confines are governed by fibroblastic reticular cells. Nat. Immunol. 15, 973鈥?81 (2014).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 7.Riedel, A., Shorthouse, D., Haas, L., Hall, B. A. Shields, J. Tumor-induced stromal reprogramming drives lymph node transformation. Nat. Immunol. 17, 1118鈥?127 (2016).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 8.Tamura, N. et al. Tumor histology in lymph vessels and lymph nodes for the accurate prediction of outcome among breast cancer patients treated with neoadjuvant chemotherapy. Cancer Sci. 100, 1823鈥?833 (2009).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 9.Estes, J. D. Pathobiology of HIV/SIV-associated changes in secondary lymphoid tissues. Immunol. Rev. 254, 65鈥?7 (2013).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 10.Zeng, M. et al. Cumulative mechanisms of lymphoid tissue fibrosis and T cell depletion in HIV-1 and SIV infections. J. Clin. Invest. 121, 998鈥?008 (2011).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 11.Lu, T. T. Browning, J. L. Role of the lymphotoxin/LIGHT system in the development and maintenance of reticular networks and vasculature in lymphoid tissues. Front Immunol. 5, 47 (2014).PubMed聽 PubMed Central聽Google Scholar聽 12.Benezech, C. et al. Lymphotoxin-beta receptor signaling through NF-kappaB2-RelB pathway reprograms adipocyte precursors as lymph node stromal cells. Immunity 37, 721鈥?34 (2012).CAS聽 PubMed聽 Article聽Google Scholar聽 13.Buckley, C. D., Barone, F., Nayar, S., Benezech, C. Caamano, J. Stromal cells in chronic inflammation and tertiary lymphoid organ formation. Annu Rev. Immunol. 33, 715鈥?45 (2015).CAS聽 PubMed聽 Article聽Google Scholar聽 14.Barone, F. et al. Stromal fibroblasts in tertiary lymphoid structures: a novel target in chronic inflammation. Front Immunol. 7, 477 (2016).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 15.Majumder, S. et al. IL-17 metabolically reprograms activated fibroblastic reticular cells for proliferation and survival. Nat. Immunol. 20, 534鈥?45 (2019).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 16.van de Pavert, S. A. et al. Chemokine CXCL13 is essential for lymph node initiation and is induced by retinoic acid and neuronal stimulation. Nat. Immunol. 10, 1193鈥?199 (2009).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 17.Kong, Y. Y. et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 397, 315鈥?23 (1999).ADS聽 CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 18.Kim, D. et al. Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE. J. Exp. Med. 192, 1467鈥?478 (2000).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 19.Benezech, C. et al. Ontogeny of stromal organizer cells during lymph node development. J. Immunol. 184, 4521鈥?530 (2010).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 20.Chai, Q. et al. Maturation of lymph node fibroblastic reticular cells from myofibroblastic precursors is critical for antiviral immunity. Immunity 38, 1013鈥?024 (2013).CAS聽 PubMed聽 Article聽Google Scholar聽 21.Onder, L. et al. Lymphatic endothelial cells control initiation of lymph node organogenesis. Immunity 47, 80鈥?2 e84 (2017).CAS聽 PubMed聽 Article聽Google Scholar聽 22.Yu, F. X., Zhao, B. Guan, K. L. Hippo pathway in organ size control, tissue homeostasis, and cancer. Cell 163, 811鈥?28 (2015).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 23.Halder, G., Dupont, S. Piccolo, S. Transduction of mechanical and cytoskeletal cues by YAP and TAZ. Nat. Rev. Mol. Cell Biol. 13, 591鈥?00 (2012).CAS聽 PubMed聽 Article聽Google Scholar聽 24.Irvine, K. D. Harvey, K. F. Control of organ growth by patterning and hippo signaling in Drosophila. Cold Spring Harb. Perspect. Biol. 7, a019224 (2015).25.Piccolo, S., Dupont, S. Cordenonsi, M. The biology of YAP/TAZ: hippo signaling and beyond. Physiol. Rev. 94, 1287鈥?312 (2014).CAS聽 PubMed聽 Article聽Google Scholar聽 26.Cho, H. et al. YAP and TAZ negatively regulate Prox1 during developmental and pathologic lymphangiogenesis. Circ. Res. 124,聽225鈥?42聽(2019).27.Xin, M. et al. Regulation of insulin-like growth factor signaling by Yap governs cardiomyocyte proliferation and embryonic heart size. Sci. Signal 4, ra70 (2011).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 28.Xin, M. et al. Hippo pathway effector Yap promotes cardiac regeneration. Proc. Natl Acad. Sci. USA 110, 13839鈥?3844 (2013).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 29.Heallen, T. et al. Hippo pathway inhibits Wnt signaling to restrain cardiomyocyte proliferation and heart size. Science 332, 458鈥?61 (2011).ADS聽 CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 30.Kim, M. et al. cAMP/PKA signalling reinforces the LATS-YAP pathway to fully suppress YAP in response to actin cytoskeletal changes. EMBO J. 32, 1543鈥?555 (2013).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 31.Hiruma, K., Hirsch, R., Patchen, M., Bluestone, J. A. Gress, R. E. Effects of anti-CD3 monoclonal antibody on engraftment of T-cell-depleted bone marrow allografts in mice: host T-cell suppression, growth factors, and space. Blood 79, 3050鈥?058 (1992).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 32.Loubaki, L., Tremblay, T. Bazin, R. In vivo depletion of leukocytes and platelets following injection of T cell-specific antibodies into mice. J. Immunol. Methods 393, 38鈥?4 (2013).CAS聽 PubMed聽 Article聽Google Scholar聽 33.Xiao, Y. et al. Hippo signaling plays an essential role in cell state transitions during cardiac fibroblast development. Dev. Cell 45, 153鈥?69 e156 (2018).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 34.dos Santos, G. et al. Vimentin regulates activation of the NLRP3 inflammasome. Nat. Commun. 6, 6574 (2015).ADS聽 PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 35.Cildir, G., Low, K. C. Tergaonkar, V. Noncanonical NF-kappaB signaling in health and disease. Trends Mol. Med. 22, 414鈥?29 (2016).CAS聽 PubMed聽 Article聽Google Scholar聽 36.Vallabhapurapu, S. Karin, M. Regulation and function of NF-kappaB transcription factors in the immune system. Annu Rev. Immunol. 27, 693鈥?33 (2009).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 37.Varelas, X. The Hippo pathway effectors TAZ and YAP in development, homeostasis and disease. Development 141, 1614鈥?626 (2014).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 38.Barrallo-Gimeno, A. Nieto, M. A. Evolutionary history of the Snail/Scratch superfamily. Trends Genet. 25, 248鈥?52 (2009).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 39.Tang, Y., Feinberg, T., Keller, E. T., Li, X. Y. Weiss, S. J. Snail/Slug binding interactions with YAP/TAZ control skeletal stem cell self-renewal and differentiation. Nat. Cell Biol. 18, 917鈥?29 (2016).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 40.Cristancho, A. G. Lazar, M. A. Forming functional fat: a growing understanding of adipocyte differentiation. Nat. Rev. Mol. Cell Biol. 12, 722鈥?34 (2011).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 41.White, A. et al. Lymphotoxin a-dependent and -independent signals regulate stromal organizer cell homeostasis during lymph node organogenesis. Blood 110, 1950鈥?959 (2007).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 42.Hong, J. H. et al. TAZ, a transcriptional modulator of mesenchymal stem cell differentiation. Science 309, 1074鈥?078 (2005).ADS聽 CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 43.Cheng, H. W. et al. Origin and differentiation trajectories of fibroblastic reticular cells in the splenic white pulp. Nat. Commun. 10, 1739 (2019).ADS聽 PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 44.Youn, D. Y. et al. Bis deficiency results in early lethality with metabolic deterioration and involution of spleen and thymus. Am. J. Physiol. Endocrinol. Metab. 295, E1349鈥揈1357 (2008).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 45.Sheikh, A. Q., Misra, A., Rosas, I. O., Adams, R. H. Greif, D. M. Smooth muscle cell progenitors are primed to muscularize in pulmonary hypertension. Sci. Transl. Med. 7, 308ra159 (2015).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 46.Novkovic, M. et al. Topological small-world organization of the fibroblastic reticular cell network determines lymph node functionality. PLoS Biol. 14, e1002515 (2016).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 47.Drumea-Mirancea, M. et al. Characterization of a conduit system containing laminin-5 in the human thymus: a potential transport system for small molecules. J. Cell Sci. 119, 1396鈥?405 (2006).CAS聽 PubMed聽 Article聽 PubMed Central聽Google Scholar聽 48.Fletcher, A. L. et al. Lymph node fibroblastic reticular cells directly present peripheral tissue antigen under steady-state and inflammatory conditions. J. Exp. Med. 207, 689鈥?97 (2010).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 49.Buenrostro, J. D., Giresi, P. G., Zaba, L. C., Chang, H. Y. Greenleaf, W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position. Nat. Methods 10, 1213鈥?218 (2013).CAS聽 PubMed聽 PubMed Central聽 Article聽Google Scholar聽 50.Knoblich, K. et al. The human lymph node microenvironment unilaterally regulates T-cell activation and differentiation. PLoS Biol. 16, e2005046 (2018).PubMed聽 PubMed Central聽 Article聽 CAS聽Google Scholar聽 51.Jeong, S. H. et al. Hippo-mediated suppression of IRS2/AKT signaling prevents hepatic steatosis and liver cancer. J. Clin. Investig. 128, 1010鈥?025 (2018).PubMed聽 Article聽Google Scholar聽 52.He, T. C. et al. A simplified system for generating recombinant adenoviruses. Proc. Natl Acad. Sci. USA 95, 2509鈥?514 (1998).ADS聽 CAS聽 PubMed聽 Article聽Google Scholar聽 Download referencesAcknowledgementsWe thank Hyun Tae Kim for technical assistance. S.Y.C. (2018R1D1A1B07044974) and H.B. (NRF- 2014-Fostering Core Leaders of the Future Basic Science Program/Global Ph.D. Fellowship Program Grant) are supported by the National Research Foundation of Korea (NRF) grant funded by the Korean government. This study was supported by the Institute for Basic Science (IBS-R025-D1, G.Y.K.) funded by the Ministry of Science, ICT and Future Planning, Korea and the Human Frontiers Science Program (RGP0034/2016, B.L. and G.Y.K.).Author informationAuthor notesThese authors contributed equally: Sung Yong Choi, Hosung Bae, Sun-Hye Jeong.AffiliationsCenter for Vascular Research, Institute for Basic Science (IBS), Daejeon, 34141, Republic of KoreaSung Yong Choi,聽Intae Park,聽Hyunsoo Cho,聽Seon Pyo Hong,聽Choong-kun Lee,聽Jin-Sung Park,聽Sang Heon Suh,聽Jeongwoon Choi,聽Myung Jin Yang,聽Joo-Hye Song聽 聽Gou Young KohGraduate School of Medical Science and Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, 34141, Republic of KoreaSung Yong Choi,聽Choong-kun Lee,聽Sang Heon Suh聽 聽Gou Young KohBiomedical Science and Engineering Interdisciplinary Program, KAIST, Daejeon, 34141, Republic of KoreaHosung Bae,聽Jeongwoon Choi,聽Myung Jin Yang聽 聽Gou Young KohDepartment of Biological Science, KAIST, Daejeon, 34141, Republic of KoreaSun-Hye Jeong,聽Da-Hye Lee聽 聽Dae-Sik LimDepartment of Otorhinolaryngology, Ajou University School of Medicine, Suwon, 16499, Republic of KoreaJeon Yeob JangInstitute of Immunobiology, Kantossipital St. Gallen, St. Gallen, 9007, SwitzerlandLucas Onder聽 聽Burkhard LudewigDepartment of Otorhinolaryngology - Head and Neck Surgery, Dankook University College of Medicine, Cheonan, 31116, Republic of KoreaJeong Hwan MoonDepartment of Otorhinolaryngology - Head and Neck Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, 06351, Republic of KoreaHan-Sin JeongDepartment of Tissue Morphogenesis, Max Planck Institute for Molecular Biomedicine, and Faculty of Medicine, University of M眉nster, M眉nster, MD-48149, GermanyRalf H. AdamsDepartment of Pathology, Chungnam National University School of Medicine, Daejeon, 35015, Republic of KoreaJin-Man KimAuthorsSung Yong ChoiView author publicationsYou can also search for this author in PubMed聽Google ScholarHosung BaeView author publicationsYou can also search for this author in PubMed聽Google ScholarSun-Hye JeongView author publicationsYou can also search for this author in PubMed聽Google ScholarIntae ParkView author publicationsYou can also search for this author in PubMed聽Google ScholarHyunsoo ChoView author publicationsYou can also search for this author in PubMed聽Google ScholarSeon Pyo HongView author publicationsYou can also search for this author in PubMed聽Google ScholarDa-Hye LeeView author publicationsYou can also search for this author in PubMed聽Google ScholarChoong-kun LeeView author publicationsYou can also search for this author in PubMed聽Google ScholarJin-Sung ParkView author publicationsYou can also search for this author in PubMed聽Google ScholarSang Heon SuhView author publicationsYou can also search for this author in PubMed聽Google ScholarJeongwoon ChoiView author publicationsYou can also search for this author in PubMed聽Google ScholarMyung Jin YangView author publicationsYou can also search for this author in PubMed聽Google ScholarJeon Yeob JangView author publicationsYou can also search for this author in PubMed聽Google ScholarLucas OnderView author publicationsYou can also search for this author in PubMed聽Google ScholarJeong Hwan MoonView author publicationsYou can also search for this author in PubMed聽Google ScholarHan-Sin JeongView author publicationsYou can also search for this author in PubMed聽Google ScholarRalf H. AdamsView author publicationsYou can also search for this author in PubMed聽Google ScholarJin-Man KimView author publicationsYou can also search for this author in PubMed聽Google ScholarBurkhard LudewigView author publicationsYou can also search for this author in PubMed聽Google ScholarJoo-Hye SongView author publicationsYou can also search for this author in PubMed聽Google ScholarDae-Sik LimView author publicationsYou can also search for this author in PubMed聽Google ScholarGou Young KohView author publicationsYou can also search for this author in PubMed聽Google ScholarContributionsS.Y.C., H.B., and S.H.J. designed and performed the experiments, analyzed and interpreted the data; S.Y.C.,聽H.B., S.H.J., I.P., H.C., and G.Y.K. wrote and edited the paper; SPH contributed to in vitro experiments; D.H.L. and M.J.Y. performed ATAC sequencing; C.K.L. and S.H.S. contributed to in vivo experiments; J.S.P. and J.C. participated in manuscript preparation; J.Y.J., J.H.M., H.S.J., and J.M.K. provided human samples; L.O., R.A., and B.L. provided the mice and critical comments on this study; J.H.S. performed immunological experiments; and D.S.L. and G.Y.K. directed and supervised the project.Corresponding authorsCorrespondence to Dae-Sik Lim or Gou Young Koh.Ethics declarations Competing interests The authors declare no competing interests. Additional informationPeer review information Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work.Publisher鈥檚 note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Supplementary information CommentsBy submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Sign up for the Nature Briefing newsletter 鈥?what matters in science, free to your inbox daily.