Differential Roles of LTβR in Endothelial Cell Subsets for Lymph Node Organogenesis and Maturation

Zhongnan Wang, Qian Chai and Mingzhao Zhu
J Immunol July 1, 2018, 201 (1) 69-76; DOI: https://doi.org/10.4049/jimmunol.1701080

Abstract

Cellular cross-talk mediated by lymphotoxin αβ–lymphotoxin β receptor (LTβR) signaling plays a critical role in lymph node (LN) development. Although the major role of LTβR signaling has long been considered to occur in mesenchymal lymphoid tissue organizer cells, a recent study using a VE-cadherincreLtbrfl/fl mouse model suggested that endothelial LTβR signaling contributes to the formation of LNs. However, the detailed roles of LTβR in different endothelial cells (ECs) in LN development remain unknown. Using various cre transgenic mouse models (Tekcre, a strain targeting ECs, and Lyve1cre, mainly targeting lymphatic ECs), we observed that specific LTβR ablation in Tekcre+ or Lyve1cre+ cells is not required for LN formation. Moreover, double-cre–mediated LTβR depletion does not interrupt LN formation. Nevertheless, TekcreLtbrfl/fl mice exhibit reduced lymphoid tissue inducer cell accumulation at the LN anlagen and impaired LN maturation. Interestingly, a subset of ECs (VE-cadherin+Tekcre-low/neg ECs) was found to be enriched in transcripts related to hematopoietic cell recruitment and transendothelial migration, resembling LN high ECs in adult animals. Furthermore, endothelial Tek was observed to negatively regulate hematopoietic cell transmigration. Taken together, our data suggest that although Tekcre+ endothelial LTβR is required for the accumulation of hematopoietic cells and full LN maturation, LTβR in VE-cadherin+Tekcre-low/neg ECs in embryos might represent a critical portal-determining factor for LN formation.

Introduction

To achieve specialized and efficient body protection, mammals have evolved a complex immune system with various organs and cells. Lymph nodes (LNs) are secondary lymphoid organs that develop as sentries at particular junctions of lymphatic and blood vessels throughout the body (1–3). They provide an effective microenvironment for Ag presentation to lymphocytes and induction of adaptive immune responses (4–6).

LN development begins during the embryonic stage, and lymphotoxin (LT) αβ–LT β receptor (LTβR) signaling plays a key role in LN development in utero (1, 7–11). Mice that are deficient in LTαβ or LTβR exhibit defective LN organogenesis or formation (12–14). LN development is a temporally and spatially highly organized process involving complex cellular and molecular cross-talk (8–10, 15). LN development is initiated with Prox1-dependent cardinal vein metamorphosis into lymphatic endothelial cells (LECs) and formation of the lymph sac, which is the LN primordial tissue, at embryonic day (E) 9 (16). Subsequently, the first hematopoietic lymphoid tissue inducer (LTi) cells are recruited to LN anlagen in a retinoic acid receptor–dependent, LTβR signaling–independent manner (17–19). An LEC-specific receptor activator for NF-κB (RANK) signal appears to play a critical role in LTi cell retention in LN anlagen (20). LTi cells are then activated to upregulate LTαβ by mesenchymal stromal cell–derived RANK ligand or IL-7 (21, 22). The cross-talk between LTαβ-expressing LTi cells and LTβR-expressing lymphoid tissue organizer (LTo) cells subsequently forms a positive feedback loop to promote further differentiation of LTo cells and production of chemokines, adhesion molecules, and cytokines, which support the attraction, retention, or activation of more hematopoietic cells, including LTi cells, leading to LN formation and further growth/maturation (18, 21–23). Therefore, in the process of LN development, LTβR has long been considered to play a role in mesenchymal LTo cells (9, 15, 18, 19, 24).

In addition to mesenchymal stromal cells, LTβR is also expressed in LECs and blood endothelial cells (BECs) (11, 25). Whether LTβR signaling directly regulates endothelial cells (ECs) in LN development remained hypothetical for quite some time (26). Recently, Ludewig and colleagues (27) have directly addressed this issue by specifically ablating LTβR expression in ECs using VE-cadherincreLtbrfl/fl mice. These authors observed 20–45% loss of peripheral LNs in these mice, suggesting an important role for endothelial-expressing LTβR signaling in LN development. However, the details of how endothelial LTβR signaling regulates LN development are unclear.

To address this question, we separately crossed Tekcre mice (28) and Lyve1cre mice (29) with Ltbrfl/fl mice (30). Unexpectedly, we found that the deficiency of LTβR in Tekcre+ ECs and Lyve1cre+ ECs, either alone or in combination, could not abrogate LN formation. Instead, we observed that LTβR signaling on Tekcre+ ECs, but not on Lyve1cre+ ECs, is critical for LN growth/maturation. More interestingly, a minor population of VE-cadherin+Tekcre-low/neg ECs present in the embryonic stage was indicated to be the critical portal cells for endothelial LTβR-mediated LN formation. Furthermore, the embryonic VE-cadherin+Tekcre-low/neg ECs resemble high ECs (HECs) in the mature LNs of adult mice. Endothelial Tek negatively regulates the transmigration of hematopoietic cells.

Materials and Methods

Mice

C57BL/6-Ly5.1 mice, B6;129P2-Lyve1tm1.1 (EGFP/cre)Cys/J mice, and B6.129 × 1-Gt (ROSA) 26Sortm1Hjf (Rosa-EYFP) mice were purchased from The Jackson Laboratory. B6.Cg-Tg (Tek-cre) 12Flv/J mice were obtained from the Nanjing Biomedical Research Institute. Ltbrfl/fl mice were obtained from Prof. Y.-X. Fu (University of Texas Southwestern Medical Center). All mice were maintained under specific pathogen-free conditions in individually ventilated cages. All animal experiments were performed in compliance with the guidelines for laboratory animals and were approved by the institutional biomedical research ethics committee of the Institute of Biophysics, Chinese Academy of Sciences.

Stromal cell preparation

Peripheral LNs or embryonic tissues were dissected into small pieces and digested in 24-well dishes containing RPMI 1640 medium with 0.2 mg/ml collagenase I (Sigma-Aldrich), 1 U/ml dispase I (Corning), 0.06 mg/ml DNase I (Roche), and 2% FBS for 1 h at 37°C. The organ fragments were subjected to pipetting every 20 min during incubation. The cell suspensions were subsequently washed with cold PBS containing 5% FBS and filtered through a 70-μm cell strainer (Biologix Group), followed by centrifugation at 1500 rpm for 5 min. The cell pellets were resuspended to obtain single-cell suspensions.

Flow cytometry and cell sorting

Single-cell suspensions were incubated on ice for 20 min in PBS containing 2% FBS and stained with the following fluorescently labeled Abs: anti-CD45 (eBioscience), anti-podoplanin (PDPN) (eBioscience), anti-CD31 (eBioscience), anti-LTβR (eBioscience), anti-CD45.1 (BioLegend), anti–VE-cadherin (eBioscience), anti-CD4 (eBioscience), anti–Lyve-1 (eBioscience), and PNAd (BioLegend). Unconjugated Abs were detected with the following secondary Ab: APC-Cy7–conjugated streptavidin (BioLegend). Flow cytometry data were acquired with an LSRFortessa (BD Biosciences) cell analyzer and analyzed with FlowJo software (Tree Star). Cells were sorted using a BD FACSAria III system.

Immunofluorescence microscopy

For LN anlagen staining, the entire LN anlage was dissected and fixed in 4% paraformaldehyde overnight before blocking and fluorescent staining with the following Abs: anti-CD31 (MEC13.3) (eBioscience), anti-CD4 (GK1.5) (eBioscience), anti-EYFP (Clontech), anti–Lyve-1 (Aly7), and anti–VE-cadherin (BV13). Unconjugated Abs were detected with the following secondary Abs: Alexa Fluor 488–conjugated anti-rabbit IgG (Jackson ImmunoResearch) and tetramethylrhodamine isothiocyanate–conjugated streptavidin (Jackson ImmunoResearch). Microscopy analysis was performed with a confocal microscope (LSM-710; Carl Zeiss), and the obtained images were processed with ZEN 2010 software (Carl Zeiss).

Quantitative real-time PCR

LN anlagen and different subsets of ECs were isolated from embryos, and RNA was extracted using TRIzol (Invitrogen) or the RNeasy Plus Micro Kit (Qiagen) following the manufacturer’s protocol. cDNA was prepared with a cDNA Reverse Transcription Kit (TransGen). Gene expression was quantified using the following primers: β-actin-forward (F): 5′-ACACCCGCCACCAGTTCGC-3′; β-actin-reverse (R): 5′-ATGGGGTACTTCAGGGTCAGGGTCAGGATA-3′; Ccl21-F: 5′-AGACTCAGGAGCCCAAAGCA-3′; Ccl21-R: 5′-GTTGAAGCAGGGCAAGGGT-3′; Ccl19-F: 5′-ATGCGGAAGACTGCTGCCT-3′; Ccl19-R: 5′-GGCTTTCACGATGTTCCCAG-3′; Madcam1-F: 5′-CCTGGCCCTAGTACCCTACC-3′; Madcam1-R: 5′-CCGTACAGAGAGGATACTGCTG-3′; Ch25h-F: 5′-TCACCAGAACTCGTCCTCCT-3′; Ch25h-R: 5′-AAGAAGGTCAGCGAAAGCAG-3′; Sphk1-F: 5′-GATGCATGAGGTGGTGAATG-3′; Sphk1-R: 5′-TGCTCGTACCCAGCATAGTG-3′; Rank-F: 5′-TCCTGGGCTTCTTCTCAGAT-3′; Rank-R: 5′-CACATCTGATTCCGTTGTCC-3′; Cxcl13-F: 5′-GCACAGCAACGCTGCTTCT-3′; Cxcl13-R: 5′-TCTTTGAACCATTTGGCAGC-3′; Tek-F: 5′-GGAGCCGCGGACTGACTAC-3′; and Tek-R: 5′-GGGAGTCCGATAGACGCTGT-3′. Quantitative real-time PCR was performed using SYBR Premix Ex Taq mix (Takara), and the reactions were run in a real-time PCR system (7500; Applied Biosystems). Relative mRNA expression levels were calculated using 7500 software v2.0.6 (Applied Biosystems).

Transwell migration assay

In the upper layer of a Transwell filter (Corning), 1 × 104 MS1 or MS1Tek−/− cells were cultured with 700 μl of 10% FBS DMEM containing angiopoietin-1 (0.4 μg/ml; Sino Biological) or all-trans retinoic acid (atRA) (5 ng/ml; Sigma-Aldrich) or with the medium alone as a control. After 24 h, 100 μl of a 1 × 106 lymphocyte suspension was prepared to replace the medium in the Transwell insert, and 600 μl of 10% FBS DMEM with 20 ng/ml CCL19 chemokine was prepared to replace the medium below the cell-permeable membrane. After 6 h, migrated lymphocytes were collected to perform flow cytometry for cell counting.

Construction of the MS1Tek−/− cell line

The CRISPR/Cas9 system was used to construct the MS1Tek−/− cell line. The single-guide RNA (sgRNA) oligo (sequence: 5′-GTAGGGAATTGATCAAGATC-3′), corresponding to a sequence located in the third exon, was synthesized and ligated into the Puro_Cas9CrisprV2 vector. The plasmid was transfected into the MS1 cell line, and the stable cell lines were screened using puromycin (2.5 μg/ml; Sigma-Aldrich). Cell lines with a deletion of thymine in the sgRNA sequence were confirmed via DNA sequencing; the sgRNA sequence of the MS1Tek−/− cell line was 5′-GTAGGGAATGATCAAGATC-3′.

Statistical analysis

All statistical analyses were assessed using the two-tailed unpaired Student t test. The p values are shown within the graphs.

Results

Tekcre+ EC-specific LTβR ablation does not impair LN organogenesis

In a previous study, endothelial LTβR ablation in VE-cadherincreLtbrfl/fl mice led to significant defects in LN formation (27). To investigate the detailed function of EC-specific LTβR signaling in LN development, we chose another common transgenic mouse model for EC targeting: Tekcre transgenic mice. Tekcre transgenic and Ltbrfl/fl mice were crossed, and the resultant TekcreLtbrfl/fl mice were examined at 4–6 wk of age. Unexpectedly, the TekcreLtbrfl/fl mice showed no abrogation of peripheral or mesenteric LN (mLN) formation (Fig. 1A–D), although the size of the LNs was smaller (Fig. 1A–C); accordingly, the total hematopoietic cell population was significantly reduced in inguinal LNs (iLNs) as well as in mLNs (Fig. 1E, 1F). These phenotypes were not due to the genomic integration or expression of cre recombinase, as Tekcre mice exhibit normal cellular compartments and total cell numbers in iLNs and mLNs, similar to those of wild-type (WT) mice (Supplemental Fig. 1). These results suggest that there is no defect in LN formation, but LN maturation is less efficient when LTβR is ablated from Tekcre+ ECs.

FIGURE 1.
FIGURE 1.

Tekcre+ EC-specific LTβR ablation does not impair LN organogenesis. (A–C) The presence of the indicated LNs was recorded in 6-wk-old TekcreLtbrfl/fl mice. Mice were injected with ink 1 d before analysis. (D) The presence of the indicated LNs was plotted as the percentage present (n > 30 mice). (E and F) LN cellularity was quantified through cell counting with single-cell suspensions from relative LNs (mean ± SEM; n = 9 mice from three independent experiments). (G) LN cell suspensions from pooled peripheral LNs of 6-wk-old TekcreRosa26-eyfp mice were prepared and analyzed via flow cytometry for CD45− stromal compartments. (H) Percentage of EYFP expression in PDPN−CD31+ BECs, PDPN+CD31+ LECs, PDPN+CD31− fibroblastic reticular cells (FRCs), PDPN−CD31− double-negative cells (DN), and CD45+ hematopoietic cells (mean ± SEM; n = 6 mice from two independent experiments). (I–L) Representative LTβR expression patterns in stromal cells of TekcreLtbrfl/fl (red) and Ltbrfl/fl (black) mice determined via flow cytometry. Isotype control staining is shown in gray. (M) LTβR expression statistics, presented as the mean fluorescence intensity (MFI) (mean ± SEM; n = 6 mice from two independent experiments). (N) Reduction rate of LTβR expression in BECs and LECs (n = 6 mice from two independent experiments).

This discrepant result led us to question whether the cre activity of Tekcre mice was sufficiently high to efficiently ablate LTβR expression in all ECs. Hence, we assessed cre activity and the LTβR ablation efficiency in TekcreLtbrfl/fl mice. By staining CD45− LN cells with PDPN and CD31, we were able to separate the LN stromal cells into four subsets (Fig. 1G). Crossing Tekcre mice with Rosa26-eyfp reporter mice (31) enabled the determination of cre activity via flow cytometry. Importantly, the specificity and efficiency of EC targeting were quite high, with >95% of the EYFP signal being detected in both PDPN−CD31+ BECs and PDPN+CD31+ LECs in the nonhematopoietic compartment (Fig. 1H). Furthermore, LTβR protein expression in both BECs and LECs, but not in other stromal cells, in the LNs of TekcreLtbrfl/fl mice was almost completely ablated, although the efficiency was slightly lower in LECs than in BECs (Fig. 1I–N). Therefore, in the Tekcre transgenic mice, nearly all ECs in LNs were targeted, and the LTβR gene was efficiently deleted in the ECs of TekcreLtbrfl/fl mice. Nonetheless, this specific and efficient LTβR ablation in Tekcre+ ECs did not abrogate LN formation.

Tekcre+ EC-specific LTβR ablation results in impaired LTi cell recruitment to embryonic LN anlagen

Previous analysis suggested that the Tekcre transgene in ECs begins to be expressed as early as E8.5 and reaches almost all ECs by E11.5, which is much earlier than the initiation of iLN formation (7, 8, 32). Moreover, effective cleavage of floxed genes by Tekcre in early embryogenesis (E8.5 or E9.5) has been demonstrated in the Tekcre transgenic system in previous studies (33, 34). However, the detailed transgene expression pattern and the ability of Tekcre to specifically and efficiently recombine Ltbr flox alleles in the LN anlage region of TekcreLtbrfl/fl mice remain to be verified. We next wondered whether the LTβR gene is also efficiently deleted in ECs at the embryonic stage, when LN formation begins. To assess Tekcre transgene activity during early LN development, we determined the reporter gene expression of TekcreRosa26-eyfp at E15.5, a critical time point for iLN formation (7, 8, 32). Skin samples including all peripheral LN anlagen were collected and then enzymatically digested for immunofluorescence staining and flow cytometric analysis. We found that >85% of BECs and ∼70% of LECs were targeted by the transgene (Fig. 2A, 2B). Specifically, in the iLN anlagen, confocal microscopy analysis showed that Tekcre+ cells were highly abundant in ECs and hematopoietic cells (Fig. 2C). Next, we directly assessed the LTβR ablation efficiency in TekcreLtbrfl/fl mice using the same gating strategies to separate stromal subsets (Fig. 2D). At E18.5, LTβR protein expression was reduced dramatically in BECs (∼80%) and to a lesser extent in LECs (∼45%) but was not reduced in fibroblastic reticular cells (Fig. 2E–G). These data suggested that Tekcre was expressed in the initial LN anlagen and that cre expression in BECs was capable of recombining the loxP sites within the Ltbr gene.

FIGURE 2.
FIGURE 2.

Tekcre+ EC-specific LTβR ablation results in impaired LTi cell recruitment to embryonic LN anlagen. (A) Skin samples including all peripheral LN anlagen were collected from E15.5 TekcreRosa26-eyfp embryos and then digested and analyzed. A representative dot plot analysis of CD31 and PDPN expression in CD45− stromal cells is shown. (B) Percentage of EYFP expression in stromal subsets and CD45+ hematopoietic cells in the skin of TekcreRosa26-eyfp mice (mean ± SEM; n > 3 mice from independent experiments). (C) Confocal microscopic analysis of the E15.5 iLN anlage of TekcreRosa26-eyfp mice. The entire LN anlage was stained with Abs against the indicated markers. (D) Representative dot plot analysis of skin cells of TekcreLtbrfl/fl mice on E18.5. (E) Representative LTβR expression patterns in stromal cell compartments of TekcreLtbrfl/fl (red) and Ltbrfl/fl (black) mice on E18.5, as determined by flow cytometry. Isotype control staining is shown in gray. (F) LTβR expression statistics, presented as the mean fluorescence intensity (MFI) in TekcreLtbrfl/fl (red) and Ltbrfl/fl (black) mice on E18.5 (mean ± SEM; n = 6 mice from two independent experiments). (G) Rate of the reduction of LTβR expression in skin BECs and LECs of TekcreLtbrfl/fl mice on E18.5 (n = 6 mice from two independent experiments). (H and I) Representative histological analysis of the iLN anlagen in TekcreLtbrfl/fl and control embryos on E18.5, stained with the indicated markers. (J) Statistics of LTi cell numbers in iLN anlagen based on confocal analysis (n > 3 from three independent experiments). (K) Statistics of CD4+ LTi cell numbers in iLN anlagen determined via flow cytometric analysis (n > 5 from two independent experiments). (L–N) Quantitative RT-PCR analysis of the expression of Ccl19 (L), Ccl21 (M), and Madcam1 (N) in iLN anlagen of TekcreLtbrfl/fl (red) and Ltbrfl/fl (black) mice on E18.5 (mean ± SEM; n > 6 mice analyzed in independent experiments). Scale bars, 50 μm.

Initial LTi cell recruitment is independent of LTβR signaling, and the subsequent LTαβ–LTβR interaction between LTi cells and mesenchymal LTo cells in the LN anlage region leads to LTo cell maturation and the recruitment of more LTi cells and other hematopoietic cells, resulting in LN maturation (19, 35). Whether endothelial LTβR is required for LTi cell recruitment at this later developmental stage for further LN maturation is not yet clear. We therefore addressed this question in TekcreLtbrfl/fl mice. At E18.5, confocal microscopy analysis showed significant reduction of CD4+ LTi cells in the iLN anlagen of TekcreLtbrfl/fl mice compared with control mice (Fig. 2H–J). This reduction was further confirmed through flow cytometric analysis (Fig. 2K). Reduction trends were also observed in other subsets of hematopoietic cells, such as B220+ and CD11c+ cells, although statistical significance was not reached (data not shown). Moreover, the expression of the chemokines CCL19 and CCL21 and the adhesion molecule MADCAM-1 in the E18.5 LN anlagen was significantly reduced in TekcreLtbrfl/fl mice (Fig. 2L–N). Taken together, these data suggest that although the ablation of LTβR on Tekcre+ ECs does not interfere with initial LN formation, it is required for the recruitment of LTi cells and other hematopoietic cells in a later phase for LN full maturation.

No impact of Lyve1cre+ EC-specific LTβR ablation on LN development

The initiation of LN formation starts from lymph sacs. Despite the coordinated development of lymphatics and LN anlagen (36, 37), the importance of the lymphatic vascular system in LN ontogeny has not been clearly elucidated. We therefore used Lyve1cre transgenic mice for specific deletion of LTβR in LECs (29). Lyve1creLtbrfl/fl mice all developed LNs normally (Fig. 3A). In addition, the total cell numbers in iLNs and mLNs were comparable to those in the control mice (Fig. 3B, 3C). These results are consistent with a recent study published during the course of our work. In that study, Onder et al. (20) observed normal LN development using the same strategy (Lyve1creLtbrfl/fl mice). Therefore, a single LTβR deletion in LECs alone appears insufficient to impact LN formation.

FIGURE 3.
FIGURE 3.

LN organogenesis is not affected in Lyve1creLtbrfl/fl or TekcreLyve1creLtbrfl/fl mice. (A) The presence of the indicated LNs in Lyve1creLtbrfl/fl mice was plotted as the percent present (n = 10 mice). (B and C) LN cellularity was quantified through cell counting using single-cell suspensions from relative LNs (mean ± SEM; n = 4 mice from two independent experiments). (D) The presence of the indicated LNs in TekcreLyve1creLtbrfl/fl mice was plotted as the percentage present (n = 4 mice). (E and F) LN cellularity was quantified with cell counting using single-cell suspensions from relative LNs (mean ± SEM; n = 4 mice from two independent experiments).

To determine whether LTβR signaling in Tekcre+ and Lyve1cre+ ECs may cooperate in LN formation, we generated TekcreLyve1creLtbrfl/fl mice. Surprisingly, all LNs still developed normally (Fig. 3D). The total cell numbers in iLNs and mLNs were reduced to a degree comparable to that in TekcreLtbrfl/fl mice (Fig. 3E, 3F), further indicating no contributing role of Lyve1cre+ EC-specific LTβR in LN formation or maturation. Taken together, these data provide compelling evidence that LTβR in Tekcre+/Lyve1cre+ ECs does not play a role in LN formation; the impairment of LN maturation is likely due to Tekcre+ EC-specific LTβR ablation.

VE-cadherin+Tekcre-low/neg ECs are characteristic of transendothelial migration

The seemingly conflicting results between our findings and the previous study suggested that there might be a minor EC subset targeted by VE-cadherincre but not by Tekcre that is essential for LN formation. To address this issue, we first determined whether all VE-cadherin+ ECs were targeted by Tekcre through flow cytometric analysis of TekcreRosa26-eyfp mice at embryonic stage E15.5. Indeed, not all VE-cadherin+ ECs were EYFP positive (Fig. 4A). We then wondered whether these VE-cadherin+Tekcre-low/neg ECs are more important for hematopoietic cell transendothelial migration compared with VE-cadherin+Tek+ ECs and may therefore represent an embryonic counterpart of HECs in adult LNs. First, most VE-cadherin+Tekcre-low/neg ECs were found to be LYVE-1 negative (Fig. 4B), suggesting they are likely BECs but not LECs. Next, we sorted VE-cadherin+Tekcre-low/neg and VE-cadherin+Tek+ ECs by flow cytometry and analyzed the expression of several functional genes for cell migration, most of which are also signature genes of HECs versus capillary ECs (38). Madcam1 is especially highly expressed in the HECs of adult Peyer’s patches and mLNs, recruiting hematopoietic cells such as CD4+CD3− LTi cells via integrin α4β7 (38, 39). Cxcl13 and Ccl21 are important chemokines that attract LTi cells to LN anlagen (17). Ch25h is critical for attracting LTi-like type 3 innate lymphoid cells (ILC3) to isolated lymphoid follicles of the colon (40). Spkh1, an essential sphingosine kinase responsible for S1P generation, has also been implicated in HEC function (41, 42). Supporting our hypothesis, the expression levels of Madcam1, Cxcl13, Ccl21, Ch25h, and Sphk1 were all significantly higher in VE-cadherin+Tekcre-low/neg ECs (Fig. 4C–G). Therefore, these data suggest that VE-cadherin+Tekcre-low/neg ECs in the embryonic stage might be particularly conditioned for LTi cell recruitment, resembling HECs in the adult animals.

FIGURE 4.
FIGURE 4.

VE-cadherin+Tekcre-low/neg ECs are characteristic of transendothelial migration. (A) Representative dot plot analysis of embryonic ECs of TekcreRosa26-eyfp mice on E15.5. (B) Representative dot plot analysis of Lyve1 expression in embryonic ECs of TekcreRosa26-eyfp mice on E15.5 for the VE-cadherin+Tekcre-low/neg EC subset. (C–G) Quantitative RT-PCR analysis of the expression of Madcam1 (C), Cxcl13 (D), Ccl21 (E), Ch25h (F), and Sphk1 (G) (mean ± SEM; n = 6 mice analyzed in independent experiments).

Endothelial Tek negatively regulates transendothelial migration

Angiopoietin-1/2 (Ang-1/2)–Tie2 (Tek) signaling is important for EC angiogenesis and homeostasis (43). The reduced expression of Tek in a subset of embryonic ECs with a potentially higher transmigration capability prompted us to investigate whether this is a general mechanism. Because high endothelial venules (HEVs) are the major cell migration portals in adult LNs, we wondered whether HECs exhibit similar downregulation of Tek expression to VE-cadherin+Tekcre-low/neg ECs in embryonic stages. HECs (CD31+PDPN−PNAd+) and non-HEV capillary BECs (CD31+PDPN−PNAd−) were sorted by flow cytometry from the LNs of adult WT mice (Fig. 5A). The expression level of Tek determined by real-time PCR was significantly lower in HECs than in capillary ECs (Fig. 5B), supporting our hypothesis. To directly test whether Tek negatively regulates endothelial transmigration, we employed an in vitro transmigration assay using MS1 cells, a blood vascular EC line (44). The Tek-deficient MS1 cell line was first generated using the CRISPR/Cas9 technique. The total lymphocyte transmigration through WT and Tek-deficient MS1 cells was compared. Indeed, Tek deficiency resulted in significantly higher numbers of transmigrated lymphocytes (Fig. 5C). To exclude any potential global impact of Tek gene knockout on MS1 cells, we activated Tek signaling with Ang-1. Pretreatment of MS1 with Ang-1 for 24 h significantly inhibited lymphocyte transmigration (Fig. 5D). These data strongly suggest that endothelial Tek may actively negatively regulate transmigration. These data not only reinforce the special role of the embryonic VE-cadherin+Tekcre-low/neg ECs during LN formation but also suggest a new mechanism for HEC regulation in adult LNs.

FIGURE 5.
FIGURE 5.

Tek negatively regulates transendothelial migration. (A) Representative dot plot analysis of ECs in the iLNs of adult WT mice. (B) Quantitative RT-PCR analysis of the expression of Tek in ECs in the iLNs of adult WT mice (n = 6 mice from two independent experiments). (C and D) Number of lymphocytes migrating through the filter in a Transwell assay. (C) MS1Tek−/− cell line compared with WT MS1 cells. (D) MS1 cells treated with or without Ang-1. Representative of two independent experiments.

Discussion

The LTβR signaling pathway is critical for LN development. The function of this pathway has long been considered to reside in mesenchymal stromal cells. A recent study suggests that endothelial LTβR is also critical for LN formation (27). However, given the various subsets of ECs, the differential roles of LTβR in LN formation and maturation in different ECs remain unclear. In this study, we clarified that LEC-specific LTβR alone is not required for LN formation or maturation. Tekcre+ BECs are required for full LN maturation but not for LN formation. Instead, a minor VE-cadherin+Tekcre-low/neg EC subset is indicated to be critical for the determination of LN formation.

Different transgenic mouse models (Tekcre and VE-cadherincre mice) were used in the current study and our previous one (27), respectively. Several possibilities exist to explain the different results. One obvious possibility is the different expression times of different cre genes. VE-cadherincre was reported to be expressed as early as E7.5 (45), whereas the Tekcre transgene in ECs was found to be expressed 1 d later at E8.5 (32). However, Tekcre activity quickly reached almost all ECs by E11.5 (32), whereas for VE-cadherincre, ∼75.5% of ECs exhibited cre activity at E12.5, and close-to-full penetrance was not observed until E14.5 (45). Considering that LTβR does not appear to play an important role in LTi accumulation in the LN anlagen until approximately E16 (19, 35), it is probably less likely that the difference in the timing of the expression of these two cre lines explains the discrepancy.

The efficiency of gene deletion mediated by these two different cre lines probably also cannot easily explain this difference. Compared with the data from a previous study involving VE-cadherincreLtbrfl/fl adult mice (27), we observed even more efficient and broader LTβR ablation in adult TekcreLtbrfl/fl mice than in VE-cadherincreLtbrfl/fl mice. LTβR expression was ablated in 98% of BECs and 92% of LECs in adult TekcreLtbrfl/fl mice. In the embryonic stage, our TekcreLtbrfl/fl mice showed early cre transgene expression in the iLN anlagen at E15.5 and an 80% LTβR ablation efficiency in BECs at E18.5. Whether endothelial LTβR was ablated earlier or more efficiently in VE-cadherincreLtbrfl/fl mice was not addressed in the previous study. Further side-by-side comparison is required to test these possibilities.

In the current work, we propose an interesting hypothesis to reconcile the observed discrepancy that a subset of ECs targeted by VE-cadherincre may not be targeted by Tekcre. Indeed, a minor VE-cadherin+Tekcre-low/neg EC subset was found at E15.5. Interestingly, this subset of ECs resembles HECs in adult animals, as demonstrated by the high expression levels of several HEC signature genes (Fig. 4C–G) and shared downregulation of Tek in both subsets (Fig. 5B), which may indicate the existence of a BEC lineage of this VE-cadherin+Tekcre-low/neg EC subset. In support of this notion, VE-cadherin+Tekcre-low/neg ECs were primarily LYVE-1 negative (∼95%) (Fig. 4B). Furthermore, both the current study and the recent study by Onder et al. (20) confirmed that a single LTβR deletion in LECs (by Lyve1cre) does not impair LN formation, suggesting that the impaired LN formation observed in VE-cadherincreLtbrfl/fl mice is likely due to LTβR deficiency in BECs. Of course, the possibility cannot be excluded that the remaining ∼5% of LYVE-1–positive cells are only the functioning population in the VE-cadherin+Tekcre-low/neg EC subset targeted by VE-cadherincre but not by Tekcre or Lyve1cre. Further investigation is needed to characterize the VE-cadherin+Tekcre-low/neg EC population in more detail and to specifically target LTβR in this EC population to determine its role in LN formation. In addition, at least ∼10% of Tekcre+ ECs were spared from LTβR deletion (based on a comparison of Fig. 2B [∼90% EGFP+ cells] and Fig. 2G [∼80% LTβR reduction]). The role of this EC population in LN formation remains to be clarified.

In the recent study by Onder et al. (20), the role of other genes in LECs in LN formation was also investigated. Although a single LTβR or NIK deletion by Lyve1cre did not impair LN formation, combined LTβR and NIK deletion by Lyve1cre resulted in a minor defect in LN formation, with the LN index decreasing from 1 to ∼0.8, indicating that LTβR in LECs may play a minor role under certain special conditions. However, a single RANK deficiency in LECs resulted in dramatic defects in LN formation, with the LN index decreasing from 1 to ∼0.4. Thus, it appears that LEC-RANK plays a major role in LN formation, whereas LEC-LTβR alone is not required. In addition, their study suggests that initial LTi recruitment occurs via LECs as early as E14, whereas after E16, activation of BECs quickly occurs, facilitating accelerated LTi recruitment. Considering all of these observations together with previous findings (8), an interesting orchestrated scenario might be that LEC-RANK plays an important role in LTi cell recruitment at the initial phase, whereas BEC-LTβR is required for further LTi cell recruitment at the later stage, in which mesenchymal stromal LTβR is also involved. NIK signaling, likely downstream of RANK, acts in coordination with LTβR for BEC activation (20).

Although LTβR signaling in Tekcre+ ECs is not responsible for LN formation, the size and cellularity of LNs in TekcreLtbrfl/fl mice (but not in Lyve1creLtbrfl/fl mice) are significantly reduced in adults, suggesting an important role for Tekcre+ endothelial LTβR (likely from BECs) in full LN maturation. Moreover, we found reduced LTi cell accumulation in LN anlagen, indicating that LTβR signaling in BECs at the late stage of LN development is critical for activation of the LN vasculature and full LN maturation. A reduced LTi cell number in LN anlagen was also recently reported in VE-cadherincreLtbrfl/flNikfl/fl mice (20). However, given the broad expression of NIK in various cell types, including LTi cells, and the activity of VE-cadherincre in both hematopoietic and ECs, it is not clear whether the reduction of LTi cell numbers was due to intrinsically impaired LTi cell development or an extrinsic endothelial defect. Because LTβR is not expressed in the lymphoid cell lineage, our data provide compelling evidence that the reduced LTi cell number is likely due to impaired endothelial function. In support of this notion, the expression of Madcam1, Ccl19, and Ccl21 was downregulated in the LN anlagen of TekcreLtbrfl/fl mice at E15.5. Notably, although LTβR is unlikely to be expressed in the lymphoid cell lineage, it is expressed in myeloid cells, particularly in CD11c+ cells. Constant LTβR signaling has been found to be required for the homeostasis of CD11c+ cells (46–48). In addition, CD11c+ cells have been reported to control initial LTi cell clustering for Peyer’s patch development and HEV homeostasis (49, 50). Furthermore, Tekcre efficiently targets almost all CD45+ hematopoietic cells (Fig. 2B). Thus, one possibility is that LTβR controls full LN maturation via CD11c+ cells in TekcreLtbrfl/fl mice. To address this issue, we generated CD11ccreLtbrfl/fl mice and found that they developed normal, fully matured LNs (Supplemental Fig. 2). These data suggest that LTβR likely does not control LN maturation via CD11c+ cells.

Although the number of LTi cells is significantly reduced in the LN anlagen of TekcreLtbrfl/fl mice, LN organogenesis still occurs. This observation suggests that there may be a threshold of the LTi cell number for the determination of LN formation and, further, that Tekcre+ endothelial LTβR deficiency alone is not sufficient to overcome this threshold requirement. In fact, Eberl et al. (35) raised the threshold theory and suggested that a threshold number of LTi cells is achieved at a later stage, after E16.5, and is dependent on the LTαβ–LTβR signaling axis. Thus, VE-cadherin+Tekcre-low/neg endothelial LTβR is probably one of the determining factors controlling the LTi cell threshold and is therefore more important in determining LN formation.

Tek is an important molecule in vascular angiogenesis and homeostasis. The reduced Tek gene expression observed in some EC subsets (VE-cadherin+Tekcre-low/neg ECs in the embryonic stage and HECs in adult animals, in this case) compared with others is unlikely to represent a simple stochastic effect and is likely a programmed event. Because Tek plays important roles in stabilized endothelial junctions (51–53), reduced Tek gene expression in HECs or HEC-like cells might be an evolutionary strategy for special permission of transendothelial migration at portal sites while maintaining intact integrity of the whole blood vessel. If Tek expression is programmed, what is the controller? In fact, atRA downregulates Tek expression in ECs (54), as confirmed in our study (Supplemental Fig. 3A). Moreover, atRA pretreatment of MS1 cells significantly enhances lymphocyte transmigration (Supplemental Fig. 3B). Therefore, retinoic acid, derived mainly from neurons adjacent to the LN anlagen in embryos not only promotes LTi cell attraction via upregulating Cxcl13 expression from mesenchymal stromal cells (17) but may also promote LTi cell transendothelial migration via inhibiting Tek-mediated stabilization of EC junctions.

In summary, by utilizing different cre transgenic mouse lines, we further investigated the differential roles of LTβR in various ECs in LN formation and maturation. The different results obtained in our study lead to a novel model in which LTβR expressed in Tekcre+ ECs is important for LN maturation, whereas LTβR in Tekcre-low/neg ECs determines LN formation. Reduced Tek expression is a programmed mechanism controlling vascular endothelial function for the recruitment of hematopoietic cells.

Disclosures

The authors have no financial conflicts of interest.

Acknowledgments

We thank Dr. Yang-Xin Fu (University of Texas Southwestern Medical Center) for providing the Ltbrfl/fl mice. We are also grateful to the staff of the Core Facility of the Institute of Biophysics, Chinese Academy of Sciences for animal health care, confocal microscopy, and flow cytometry services.

Footnotes

  • This work was supported by grants from the National Natural Science Foundation of China (31570888, 81373110, and 81261130022 to M.Z. and 31400758 to Q.C.), the Ministry of Science and Technology (2015CB943400 to Q.C.), and the Youth Innovation Promotion Association of the Chinese Academy of Sciences (2016089 to Q.C.).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    atRA
    all-trans retinoic acid
    BEC
    blood endothelial cell
    E
    embryonic day
    EC
    endothelial cell
    F
    forward
    HEC
    high EC
    HEV
    high endothelial venule
    iLN
    inguinal LN
    LEC
    lymphatic endothelial cell
    LN
    lymph node
    LT
    lymphotoxin
    LTi
    lymphoid tissue inducer
    LTo
    lymphoid tissue organizer
    LTβR
    LT β receptor
    mLN
    mesenteric LN
    PDPN
    podoplanin
    R
    reverse
    RANK
    receptor activator for NF-κB
    sgRNA
    single-guide RNA
    WT
    wild-type.

  • Received July 26, 2017.
  • Accepted April 26, 2018.

References

  1. ↵
    1. Fu, Y. X.,
    2. D. D. Chaplin
    . 1999. Development and maturation of secondary lymphoid tissues. Annu. Rev. Immunol. 17: 399–433.
    OpenUrlCrossRefPubMed
    1. Drayton, D. L.,
    2. S. Liao,
    3. R. H. Mounzer,
    4. N. H. Ruddle
    . 2006. Lymphoid organ development: from ontogeny to neogenesis. Nat. Immunol. 7: 344–353.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Junt, T.,
    2. E. Scandella,
    3. B. Ludewig
    . 2008. Form follows function: lymphoid tissue microarchitecture in antimicrobial immune defence. Nat. Rev. Immunol. 8: 764–775.
    OpenUrlCrossRefPubMed
  4. ↵
    1. Qi, H.,
    2. W. Kastenmüller,
    3. R. N. Germain
    . 2014. Spatiotemporal basis of innate and adaptive immunity in secondary lymphoid tissue. Annu. Rev. Cell Dev. Biol. 30: 141–167.
    OpenUrlCrossRefPubMed
    1. Mueller, S. N.,
    2. R. N. Germain
    . 2009. Stromal cell contributions to the homeostasis and functionality of the immune system. Nat. Rev. Immunol. 9: 618–629.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Turley, S. J.,
    2. A. L. Fletcher,
    3. K. G. Elpek
    . 2010. The stromal and haematopoietic antigen-presenting cells that reside in secondary lymphoid organs. Nat. Rev. Immunol. 10: 813–825.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Rennert, P. D.,
    2. J. L. Browning,
    3. R. Mebius,
    4. F. Mackay,
    5. P. S. Hochman
    . 1996. Surface lymphotoxin alpha/beta complex is required for the development of peripheral lymphoid organs. J. Exp. Med. 184: 1999–2006.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Mebius, R. E.
    2003. Organogenesis of lymphoid tissues. [Published erratum appears in 2003 Nat. Rev. Immunol. 3: 509.] Nat. Rev. Immunol. 3: 292–303.
    OpenUrlCrossRefPubMed
  9. ↵
    1. van de Pavert, S. A.,
    2. R. E. Mebius
    . 2010. New insights into the development of lymphoid tissues. Nat. Rev. Immunol. 10: 664–674.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Randall, T. D.,
    2. D. M. Carragher,
    3. J. Rangel-Moreno
    . 2008. Development of secondary lymphoid organs. Annu. Rev. Immunol. 26: 627–650.
    OpenUrlCrossRefPubMed
  11. ↵
    1. Lu, T. T.,
    2. J. L. Browning
    . 2014. Role of the lymphotoxin/LIGHT system in the development and maintenance of reticular networks and vasculature in lymphoid tissues. Front. Immunol. 5: 47.
    OpenUrlPubMed
  12. ↵
    1. Fütterer, A.,
    2. K. Mink,
    3. A. Luz,
    4. M. H. Kosco-Vilbois,
    5. K. Pfeffer
    . 1998. The lymphotoxin beta receptor controls organogenesis and affinity maturation in peripheral lymphoid tissues. Immunity 9: 59–70.
    OpenUrlCrossRefPubMed
    1. Alimzhanov, M. B.,
    2. D. V. Kuprash,
    3. M. H. Kosco-Vilbois,
    4. A. Luz,
    5. R. L. Turetskaya,
    6. A. Tarakhovsky,
    7. K. Rajewsky,
    8. S. A. Nedospasov,
    9. K. Pfeffer
    . 1997. Abnormal development of secondary lymphoid tissues in lymphotoxin beta-deficient mice. Proc. Natl. Acad. Sci. USA 94: 9302–9307.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Koni, P. A.,
    2. R. Sacca,
    3. P. Lawton,
    4. J. L. Browning,
    5. N. H. Ruddle,
    6. R. A. Flavell
    . 1997. Distinct roles in lymphoid organogenesis for lymphotoxins α and β revealed in lymphotoxin β-deficient mice. Immunity 6: 491–500.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Blum, K. S.,
    2. R. Pabst
    . 2006. Keystones in lymph node development. J. Anat. 209: 585–595.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Srinivasan, R. S.,
    2. M. E. Dillard,
    3. O. V. Lagutin,
    4. F. J. Lin,
    5. S. Tsai,
    6. M. J. Tsai,
    7. I. M. Samokhvalov,
    8. G. Oliver
    . 2007. Lineage tracing demonstrates the venous origin of the mammalian lymphatic vasculature. Genes Dev. 21: 2422–2432.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. van de Pavert, S. A.,
    2. B. J. Olivier,
    3. G. Goverse,
    4. M. F. Vondenhoff,
    5. M. Greuter,
    6. P. Beke,
    7. K. Kusser,
    8. U. E. Höpken,
    9. M. Lipp,
    10. K. Niederreither, et al
    . 2009. Chemokine CXCL13 is essential for lymph node initiation and is induced by retinoic acid and neuronal stimulation. Nat. Immunol. 10: 1193–1199.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Bénézech, C.,
    2. A. White,
    3. E. Mader,
    4. K. Serre,
    5. S. Parnell,
    6. K. Pfeffer,
    7. C. F. Ware,
    8. G. Anderson,
    9. J. H. Caamaño
    . 2010. Ontogeny of stromal organizer cells during lymph node development. J. Immunol. 184: 4521–4530.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. White, A.,
    2. D. Carragher,
    3. S. Parnell,
    4. A. Msaki,
    5. N. Perkins,
    6. P. Lane,
    7. E. Jenkinson,
    8. G. Anderson,
    9. J. H. Caamaño
    . 2007. Lymphotoxin a-dependent and -independent signals regulate stromal organizer cell homeostasis during lymph node organogenesis. Blood 110: 1950–1959.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Onder, L.,
    2. U. Mörbe,
    3. N. Pikor,
    4. M. Novkovic,
    5. H. W. Cheng,
    6. T. Hehlgans,
    7. K. Pfeffer,
    8. B. Becher,
    9. A. Waisman,
    10. T. Rülicke, et al
    . 2017. Lymphatic endothelial cells control initiation of lymph node organogenesis. Immunity 47: 80–92.e4.
    OpenUrl
  21. ↵
    1. Yoshida, H.,
    2. A. Naito,
    3. J. Inoue,
    4. M. Satoh,
    5. S. M. Santee-Cooper,
    6. C. F. Ware,
    7. A. Togawa,
    8. S. Nishikawa,
    9. S. Nishikawa
    . 2002. Different cytokines induce surface lymphotoxin-alphabeta on IL-7 receptor-alpha cells that differentially engender lymph nodes and Peyer’s patches. Immunity 17: 823–833.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Vondenhoff, M. F.,
    2. M. Greuter,
    3. G. Goverse,
    4. D. Elewaut,
    5. P. Dewint,
    6. C. F. Ware,
    7. K. Hoorweg,
    8. G. Kraal,
    9. R. E. Mebius
    . 2009. LTbetaR signaling induces cytokine expression and up-regulates lymphangiogenic factors in lymph node anlagen. J. Immunol. 182: 5439–5445.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Honda, K.,
    2. H. Nakano,
    3. H. Yoshida,
    4. S. Nishikawa,
    5. P. Rennert,
    6. K. Ikuta,
    7. M. Tamechika,
    8. K. Yamaguchi,
    9. T. Fukumoto,
    10. T. Chiba,
    11. S. I. Nishikawa
    . 2001. Molecular basis for hematopoietic/mesenchymal interaction during initiation of Peyer’s patch organogenesis. J. Exp. Med. 193: 621–630.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Brendolan, A.,
    2. J. H. Caamaño
    . 2012. Mesenchymal cell differentiation during lymph node organogenesis. Front. Immunol. 3: 381.
    OpenUrlCrossRefPubMed
  25. ↵
    1. Zhu, M.,
    2. Y. X. Fu
    . 2011. The role of core TNF/LIGHT family members in lymph node homeostasis and remodeling. Immunol. Rev. 244: 75–84.
    OpenUrlCrossRefPubMed
  26. ↵
    1. Nishikawa, S.,
    2. K. Honda,
    3. P. Vieira,
    4. H. Yoshida
    . 2003. Organogenesis of peripheral lymphoid organs. Immunol. Rev. 195: 72–80.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Onder, L.,
    2. R. Danuser,
    3. E. Scandella,
    4. S. Firner,
    5. Q. Chai,
    6. T. Hehlgans,
    7. J. V. Stein,
    8. B. Ludewig
    . 2013. Endothelial cell-specific lymphotoxin-β receptor signaling is critical for lymph node and high endothelial venule formation. J. Exp. Med. 210: 465–473.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Koni, P. A.,
    2. S. K. Joshi,
    3. U. A. Temann,
    4. D. Olson,
    5. L. Burkly,
    6. R. A. Flavell
    . 2001. Conditional vascular cell adhesion molecule 1 deletion in mice: impaired lymphocyte migration to bone marrow. J. Exp. Med. 193: 741–754.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    1. Pham, T. H.,
    2. P. Baluk,
    3. Y. Xu,
    4. I. Grigorova,
    5. A. J. Bankovich,
    6. R. Pappu,
    7. S. R. Coughlin,
    8. D. M. McDonald,
    9. S. R. Schwab,
    10. J. G. Cyster
    . 2010. Lymphatic endothelial cell sphingosine kinase activity is required for lymphocyte egress and lymphatic patterning. J. Exp. Med. 207: 17–27.
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Wang, Y.,
    2. E. P. Koroleva,
    3. A. A. Kruglov,
    4. D. V. Kuprash,
    5. S. A. Nedospasov,
    6. Y. X. Fu,
    7. A. V. Tumanov
    . 2010. Lymphotoxin beta receptor signaling in intestinal epithelial cells orchestrates innate immune responses against mucosal bacterial infection. Immunity 32: 403–413.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Srinivas, S.,
    2. T. Watanabe,
    3. C. S. Lin,
    4. C. M. William,
    5. Y. Tanabe,
    6. T. M. Jessell,
    7. F. Costantini
    . 2001. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev. Biol. 1: 4.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Kisanuki, Y. Y.,
    2. R. E. Hammer,
    3. J. Miyazaki,
    4. S. C. Williams,
    5. J. A. Richardson,
    6. M. Yanagisawa
    . 2001. Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo. Dev. Biol. 230: 230–242.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Cattelino, A.,
    2. S. Liebner,
    3. R. Gallini,
    4. A. Zanetti,
    5. G. Balconi,
    6. A. Corsi,
    7. P. Bianco,
    8. H. Wolburg,
    9. R. Moore,
    10. B. Oreda, et al
    . 2003. The conditional inactivation of the beta-catenin gene in endothelial cells causes a defective vascular pattern and increased vascular fragility. J. Cell Biol. 162: 1111–1122.
    OpenUrlAbstract/FREE Full Text
  34. ↵
    1. Friedrich, E. B.,
    2. E. Liu,
    3. S. Sinha,
    4. S. Cook,
    5. D. S. Milstone,
    6. C. A. MacRae,
    7. M. Mariotti,
    8. P. J. Kuhlencordt,
    9. T. Force,
    10. A. Rosenzweig, et al
    . 2004. Integrin-linked kinase regulates endothelial cell survival and vascular development. Mol. Cell. Biol. 24: 8134–8144.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Eberl, G.,
    2. S. Marmon,
    3. M. J. Sunshine,
    4. P. D. Rennert,
    5. Y. Choi,
    6. D. R. Littman
    . 2004. An essential function for the nuclear receptor RORgamma(t) in the generation of fetal lymphoid tissue inducer cells. Nat. Immunol. 5: 64–73.
    OpenUrlCrossRefPubMed
  36. ↵
    1. van de Pavert, S. A.,
    2. R. E. Mebius
    . 2014. Development of secondary lymphoid organs in relation to lymphatic vasculature. Adv. Anat. Embryol. Cell Biol. 214: 81–91.
    OpenUrl
  37. ↵
    1. Lee, Y. G.,
    2. G. Y. Koh
    . 2016. Coordinated lymphangiogenesis is critical in lymph node development and maturation. Dev. Dyn. 245: 1189–1197.
    OpenUrl
  38. ↵
    1. Lee, M.,
    2. H. Kiefel,
    3. M. D. LaJevic,
    4. M. S. Macauley,
    5. H. Kawashima,
    6. E. O’Hara,
    7. J. Pan,
    8. J. C. Paulson,
    9. E. C. Butcher
    . 2014. Transcriptional programs of lymphoid tissue capillary and high endothelium reveal control mechanisms for lymphocyte homing. [Published erratum appears in 2015 Nat. Immunol. 16: 214.] Nat. Immunol. 15: 982–995.
    OpenUrlCrossRefPubMed
  39. ↵
    1. Mebius, R. E.,
    2. I. L. Schadee-Eestermans,
    3. I. L. Weissman
    . 1998. MAdCAM-1 dependent colonization of developing lymph nodes involves a unique subset of CD4+CD3- hematolymphoid cells. Cell Adhes. Commun. 6: 97–103.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Emgård, J.,
    2. H. Kammoun,
    3. B. García-Cassani,
    4. J. Chesné,
    5. S. M. Parigi,
    6. J. M. Jacob,
    7. H. W. Cheng,
    8. E. Evren,
    9. S. Das,
    10. P. Czarnewski, et al
    . 2018. Oxysterol sensing through the receptor GPR183 promotes the lymphoid-tissue-inducing function of innate lymphoid cells and colonic inflammation. Immunity 48: 120–132.e8.
    OpenUrl
  41. ↵
    1. Halin, C.,
    2. M. L. Scimone,
    3. R. Bonasio,
    4. J. M. Gauguet,
    5. T. R. Mempel,
    6. E. Quackenbush,
    7. R. L. Proia,
    8. S. Mandala,
    9. U. H. von Andrian
    . 2005. The S1P-analog FTY720 differentially modulates T-cell homing via HEV: T-cell-expressed S1P1 amplifies integrin activation in peripheral lymph nodes but not in Peyer patches. Blood 106: 1314–1322.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    1. Limaye, V.,
    2. M. A. Vadas,
    3. S. M. Pitson,
    4. J. R. Gamble
    . 2009. The effects of markedly raised intracellular sphingosine kinase-1 activity in endothelial cells. Cell. Mol. Biol. Lett. 14: 411–423.
    OpenUrlPubMed
  43. ↵
    1. Moss, A.
    2013. The angiopoietin:Tie 2 interaction: a potential target for future therapies in human vascular disease. Cytokine Growth Factor Rev. 24: 579–592.
    OpenUrlCrossRefPubMed
  44. ↵
    1. Ledgerwood, L. G.,
    2. G. Lal,
    3. N. Zhang,
    4. A. Garin,
    5. S. J. Esses,
    6. F. Ginhoux,
    7. M. Merad,
    8. H. Peche,
    9. S. A. Lira,
    10. Y. Ding, et al
    . 2008. The sphingosine 1-phosphate receptor 1 causes tissue retention by inhibiting the entry of peripheral tissue T lymphocytes into afferent lymphatics. Nat. Immunol. 9: 42–53.
    OpenUrlCrossRefPubMed
  45. ↵
    1. Alva, J. A.,
    2. A. C. Zovein,
    3. A. Monvoisin,
    4. T. Murphy,
    5. A. Salazar,
    6. N. L. Harvey,
    7. P. Carmeliet,
    8. M. L. Iruela-Arispe
    . 2006. VE-Cadherin-Cre-recombinase transgenic mouse: a tool for lineage analysis and gene deletion in endothelial cells. Dev. Dyn. 235: 759–767.
    OpenUrlCrossRefPubMed
  46. ↵
    1. De Trez, C.
    2012. Lymphotoxin-beta receptor expression and its related signaling pathways govern dendritic cell homeostasis and function. Immunobiology 217: 1250–1258.
    OpenUrl
    1. Wang, Y. G.,
    2. K. D. Kim,
    3. J. Wang,
    4. P. Yu,
    5. Y. X. Fu
    . 2005. Stimulating lymphotoxin beta receptor on the dendritic cells is critical for their homeostasis and expansion. J. Immunol. 175: 6997–7002.
    OpenUrlAbstract/FREE Full Text
  48. ↵
    1. Kabashima, K.,
    2. T. A. Banks,
    3. K. M. Ansel,
    4. T. T. Lu,
    5. C. F. Ware,
    6. J. G. Cyster
    . 2005. Intrinsic lymphotoxin-beta receptor requirement for homeostasis of lymphoid tissue dendritic cells. Immunity 22: 439–450.
    OpenUrlCrossRefPubMed
  49. ↵
    1. Veiga-Fernandes, H.,
    2. M. C. Coles,
    3. K. E. Foster,
    4. A. Patel,
    5. A. Williams,
    6. D. Natarajan,
    7. A. Barlow,
    8. V. Pachnis,
    9. D. Kioussis
    . 2007. Tyrosine kinase receptor RET is a key regulator of Peyer’s patch organogenesis. Nature 446: 547–551.
    OpenUrlCrossRefPubMed
  50. ↵
    1. Moussion, C.,
    2. J. P. Girard
    . 2011. Dendritic cells control lymphocyte entry to lymph nodes through high endothelial venules. Nature 479: 542–546.
    OpenUrlCrossRefPubMed
  51. ↵
    1. Suri, C.,
    2. P. F. Jones,
    3. S. Patan,
    4. S. Bartunkova,
    5. P. C. Maisonpierre,
    6. S. Davis,
    7. T. N. Sato,
    8. G. D. Yancopoulos
    . 1996. Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 87: 1171–1180.
    OpenUrlCrossRefPubMed
    1. Frye, M.,
    2. M. Dierkes,
    3. V. Küppers,
    4. M. Vockel,
    5. J. Tomm,
    6. D. Zeuschner,
    7. J. Rossaint,
    8. A. Zarbock,
    9. G. Y. Koh,
    10. K. Peters, et al
    . 2015. Interfering with VE-PTP stabilizes endothelial junctions in vivo via Tie-2 in the absence of VE-cadherin. J. Exp. Med. 212: 2267–2287.
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Jeansson, M.,
    2. A. Gawlik,
    3. G. Anderson,
    4. C. Li,
    5. D. Kerjaschki,
    6. M. Henkelman,
    7. S. E. Quaggin
    . 2011. Angiopoietin-1 is essential in mouse vasculature during development and in response to injury. J. Clin. Invest. 121: 2278–2289.
    OpenUrlCrossRefPubMed
  54. ↵
    1. Suzuki, Y.,
    2. Y. Komi,
    3. H. Ashino,
    4. J. Yamashita,
    5. J. Inoue,
    6. A. Yoshiki,
    7. A. Eichmann,
    8. H. Amanuma,
    9. S. Kojima
    . 2004. Retinoic acid controls blood vessel formation by modulating endothelial and mural cell interaction via suppression of Tie2 signaling in vascular progenitor cells. Blood 104: 166–169.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section