Site Loader

Thymocytes in Lyve1-CRE; S1pr1f/f mice
accumulate in the thymus
due to cell-intrinsic loss of
sphingosine-1-phosphate receptor expression

Mohammad Shahadat Hossain

Master´s thesis

University of Turku
Faculty of Medicine

Masters degree in Biomedical Imaging

Credits: 45 ECTS

1: Akira Takeda, Ph.D; University of Turku, Finland
2: Masayuki Miyasaka, MD, Ph.D; Osaka University, Japan

1: Professor
2: Professor



The originality of this thesis has been verified in accordance with the University of Turku quality assurance system using the Turnitin Originality Check service.

Faculty of Medicine

Thymocytes in Lyve1-Cre; S1pr1f/f mice accumulate in the thymus due to cell-intrinsic loss of sphingosine-1-phosphate receptor expression

Master´s thesis
March, 2018

T cell egress from the lymphoid tissues is essential for immunological homeostasis. While sphingosine-1-phosphate (S1P) produced by stromal lymphatic endothelial cells has been shown to promote lymphocyte egress via the S1P receptor S1PR1 on lymphocytes, the significance of S1PR1 signaling in the stromal lymphatic endothelial cells remains unclear. To address this issue, we developed conditional knockout mice (Lyve1-Cre; S1pr1f/f mice) where lymphatic endothelial cells are rendered null for S1pr1. In these mice, T cells were significantly reduced in secondary lymphoid tissues, and CD62L+ mature CD4 and CD8 single-positive (SP) T cells were accumulated in the medulla and failed to egress from the thymus. Using a Lyve1-Cre reporter strain in which Lyve1-linage cells expressed tdTomato fluorescent protein, we unexpectedly found that a considerable proportions of lymphocytes expressed tdTomato fluorescent protein, indicating they belonged to Lyve1-lineage. The CD4 and CD8 SP thymocytes in Lyve1-Cre; S1pr1f/f mice showed an egress-competent phenotype (HSAlow, CD62Lhigh, and Qa2high), but were CD69high and lacked S1PR1 expression. In addition, CD4 SP thymocytes were unable to migrate into the periphery after their intrathymic injection into wild-type mice. In contrast, WT thymocytes could migrate to the periphery in both S1pr1f/f and Lyve1-Cre; S1pr1f/f thymuses. These results demonstrated thymocyte egress is mediated by T cell-expressed, but not lymphatics-expressed, S1PR1 and caution against using the Lyve1-Cre system to delete genes selectively in lymphatic endothelial cells.

Keywords: sphingosine-1-phosphate, lymphocyte trafficking, thymus, Lyve1

Table of Contents

1.1. Lymphocyte trafficking in the body 2
1.2. Sphingosine-1 phosphate (S1P): synthesis and metabolism 5
1.3. S1P receptors maintain vascular integrity 8
1.4. S1P regulates lymphocyte egress from lymphoid organs 10
1.5. Spatial control of S1P generation and receptor activation in vivo 11

2. AIMS 15
3.1. Mice 16
3.2. Immunohistochemical analysis 20
3.3. Flow cytometric analysis 23
3.4. Intrathymic injection 28
3.5. Adoptive homing assays 28
3.6. Statistical analysis 33
4.1. T cells are substantially reduced in the secondary lymphoid tissues of the Lyve1-CRE S1pr1f/f mice 34
4.2. Reduction of circulating T and B cells is not due to alteration in homing properties 38
4.3. Qa-2+ CD4+ SP and CD8+ SP subsets are markedly accumulated in the thymic medulla of the Lyve1-CRE; S1pr1f/f thymus 39
4.4. The medullary SP subsets in Lyve1-CRE; S1pr1f/f mice phenotypically resemble egress-competent thymocytes in wild-type Mice but lack S1PR1 expression 43
4.5. A majority of medullary thymocytes belong to the Lyve1 lineage 43
4.6. The SP subsets in the Lyve1-CRE; S1pr1f/f thymus are egress-incompetent Lyve1-CRE; S1pr1f/f thymus 43



CMTMR (5-(and-6)-(((4-chloromethyl)benzoyl)amino) tetramethylrhodamine
CFSE Carboxyfluorescein succinimide ester
DP Double Positive
ECs Endothelial Cells
GPCR G protein-coupled receptor
HEVs High endothelial venules
LECs Lymphatic endothelial cells
LN Lymph node
LPP3 Lysophospholipid phosphatase 3
Lyve1 Lymphatic vessel endothelial hyaluronic acid receptor 1
S1P Sphingosine-1-phosphate
S1PR Sphingosine-1-phosphate receptor
Sphk Sphingosine kinase
SP Single Positive
WT Wild type

1.1. Lymphocyte trafficking in the body

Continuous lymphocyte recirculation between blood and lymphoid organs is a fundamental element of the immune system (von Andrian and Mempel, 2003). It provides lymphocytes with the ability to travel throughout the body in search of antigens that represent a constant threat to our well-being. The immune system relies on the persistent movement of lymphocytes through various anatomical sites (Cyster and Schwab, 2011). Lymphocytes are generated in the primary lymphoid organs including the bone marrow and thymus and emigrate into the blood to migrate into the secondary lymphoid organs such as lymph nodes (LNs), spleen and Peyer’s patches. Lymphocytes enter secondary lymphoid organs through high endothelial venules (HEVs), which requires several consecutives steps: lymphocyte rolling, integrin activation, adhesion and transmigration (Andrian and Mempel, 2003). Transmigrated lymphocytes survey antigens in the lymphoid organs for several hours, and they exit through efferent lymphatics to travel to other lymphoid organs. The lymphocytes are then recirculated into the blood through the thoracic ducts. Upon inflammation in the peripheral tissues, lymphocyte egress from the draining LNs is quickly shut down to maximize the possibility of encountering with antigens by rare cognate lymphocytes. Lymphocytes with a T cell receptor against the antigens form stable conjugates with antigen-presenting cells in the lymph organs during activation and proliferation. Activated effector lymphocytes emigrate from LNs, migrate to the site of infection and mediate protective functions.

Figure-1: Lymphocyte recirculation. Haematopoietic stem cells are dispersed from the bone marrow as either T-cell precursors or B cells into the bloodstream. The T cell precursor cells migrate into the thymus and they differentiate into T cells (blue colored). The T cells then migrate to the blood and secondary lymphoid tissues such as LNs and spleen, and then return to the blood via the thoracic ducts. B cells (red) migrate from bone marrow to the blood in a CXCR4-depedent manner and recirculate in the same way as T cells. In mice, 80–90% of the lymphocytes in the recirculating pool are T cells.This figure is from Miller J.F., 2011.
1.2. Sphingosine -1-phosphate (S1P): synthesis and metabolism

Sphingosine-1-phosphate (S1P) is a lipid mediator formed by the metabolism of sphingomyelin. Sphingomyelinase generates ceramide from sphingomyelin, and ceramidase produces sphingosine from ceramide. S1P is finally generated from sphingosine by the action of sphingosine kinase 1 and 2 (Sphk1/2). The abundance of intracellular S1P is regulated by S1P biosynthetic enzymes (Sphk1/2) and degradative enzymes such as S1P lyase (Spgl1) and lysophospholipid phosphatases (Ppap2a, b, c) (Blaho and Hla, 2011; Le Stunff et al, 2004; Saba and Hla, 2004). S1P generated inside the cells is then exported by a specific transporter Spinster 2 (Spns2) (Kawahara et al., 2009; Osborne et al., 2008) and acts on S1P receptors on target cells. Recent study showed that Spns2 is expressed in mouse vascular endothelial cells and is involved in generation of S1P in the plasma (Fukuhara et al, 2012).

Figure 2: S1P synthesis and metabolism. Sphingosine 1-phosphate is a lipid mediator formed by the metabolism of sphingomyelin. The levels of intracellura and extracellular S1P are regulated by both S1P biosynthetic and degradative enzymes. This figure is from Mendelson et al., 2014

Concentration of extracellular S1P is also tightly regulated by generation and degradation. A secreted form of Sphk1 from endothelial cells (ECs) plays a role in generating a gradient of S1P between vascular and extravascular compartment (Venkataraman et al, 2006). S1P is abundant in the lymph and blood, whereas the concentration is low in lymphoid organs by the action of S1P lyase and lysophospholipid phosphatase 3 (LPP3) (Schwab, 2005; Breart, 2011). This creates a gradient of S1P, which drive lymphocyte chemotaxis to the blood or lymph (Cyster and Schwab, 2012). S1P in the plasma is bound to albumin and apolipoprotein M-containing high-density lipoprotein (ApoM+ HDL) and this binding enhances the function of S1P and vascular integrity (Christoffersen et al, 2011).

1.3. S1P receptors maintain vascular integrity

Receptors of S1P were first identified in the vascular ECs as endothelial differentiation genes (Edg receptors) in the early 1990s. They are G protein coupled receptors, and five receptors have been identified so far (S1PR1-5). Among S1PR, S1PR1 is well characterized. S1PR1-deficient mice showed embryonic hemorrhage leading to intrauterine death between E12.5 and E14.5 due to a defect in vascular maturation (Liu et al, 2000). ECs in the embryo had incomplete vascular smooth muscle cell coverage. Deletion of S1P-generating enzymes Sphk1 and Sphk2 also led to embryonic lethality in mid-gestation (Mizugishi, 2005). Interestingly, S1PR1 is activated not only by ligands but also by laminar shear stress to preserve vascular integrity (Jung, 2012). Among S1P receptors, S1PR2 and S1PR3 are also expressed vascular ECs and they have been shown to work redundantly and cooperatively with S1PR1 to stabilize vascular system during development (Kono et al, 2004).
Disruption of S1PR1 solely in ECs also lead to the defects observed in the embryo globally deficient in S1PR1, suggesting that vessel coverage by smooth muscle cells is regulated by S1PR in ECs (Allende, 2003). Postnatal deletion of S1pr1 selectively in ECs using Cdh5-CreERT2; S1pr1fl/fl mice (S1pr1 ECKO) showed abnormal vascular phenotype characterized by defective vascularization, hypersprouting, increased branching, and dilated morphology in retinas. (Jung et al, 2012).Endothelial hypersprouting caused by S1PR1-deficiency has also been described in other vascular beds, such as the embryonic hindbrain, aorta, neural tube and developing limbs of mice (Gaengel et al, 2012; Ben Shoham et al, 2012); and the hindbrain vessels and the caudal vein plexus of zebrafish (Gaengel et al, 2012; Mendelson et al, 2013; Ben Shoham et al, 2012).

Figure 3: S1P receptor in vascular development. Tip cells express S1PR1 but are poorly exposed to S1P. Once Tip cells are integrated into the vasucular lumen, S1P in the blood activates S1PR1 and enhances VE-cadherin-mediated adherent junction formation, which leads to inhibition of VEGFA signaling and suppression of sprouting. This figure is from Mendelson et al., 2014.

Absent of VE-cadherin in the ECs shows retinal angiogenic hypersprouting phenotype that is similar to the S1pr1 ECKO mice (Gaengel et al, 2012). Zebrafish morpholino-based studies demonstrated the cooperation of VE-cadherin and S1PR1 for inhibiting vascular sprouting and promoting vascular stability. Furthermore, S1PR1 inhibits vascular endothelial growth facter-A induced angiogenic signaling in vitro in human umbilical vein endothelial cells (HUVECs) and in vivo in mice and zebrafish (Gaengel et al, 2012; Ben Shoham et al, 2012). Deletion of Vegfa or Hif1a, a transcription factor of Vegfa, causes a reduction of blood vessel density in the developing mouse forelimb. On the other hand, knockdown of S1pr1 leads to an increase in blood vessel density (Ben Shoham et al, 2012). Double knockdown of Hif1a and S1pr1 reduced blood vessel density of S1pr1-/- (Ben Shoham et al, 2012). These studies indicate that S1PR1 prevent excessive vascular sprouting at the cell junctions through stabilization of VE-cadherin and also through inhibition of VEGFR2 signaling and phosphorylation.

1.4. S1P regulates lymphocyte egress from lymphoid organs

An immune suppressive compound FTY720 (2-amino-2-(2-4-octylphenyl ethyl)-1, 3-propanediol) was synthesized by the chemical derivatization process of myriocin that is a sphingosine-like metabolite of the ascomycete Isaria sinclairii (Adachi et al, 1995). FTY720 shows dramatic immunosuppressive activity and prolongs survival of solid organ allograft in animal models. Although myriocin shows an immunosuppressive activity via binding to serine palmitoyl transferase and subsequently inducing T cell apoptosis, FTY720 does not bind to serine palmitoyl transferase (Chen et al, 1999) and the mechanism of immunosuppressive action has not been clear for several years. On the other hand, administration of FTY720 in in vivo lymphopenia by sequestration of lymphocytes in the spleen, LNs and Peyer’s patches (Pinschewer et al, 2000; Brinkmann et al, 2000 and Chiba et al, 1998). These results raise a hypothesis that FTY720 may accelerate lymphocyte homing in secondary lymphoid organs via upregulating responsiveness to chemokines into the lymphoid organs (Chiba et al, 1998). However, several studies showed that the sequestration of lymphocytes in secondary lymphoid organs caused by FTY720 was not dependent on the homing receptors CD62L (Bai et al, 2002), and CCR7 (Henning et al, 2001), and CCR7-ligands CCL19 and CCL21 (Henning et al, 2001). In 2002, Mandala et al elegantly showed that FTY720 is a high-affinity agonist of four of the five S1P receptors and that it induces emptying of efferent lymphatics in LNs by inhibiting lymphocyte egress into lymph (Mandala et al, 2002). Furthermore, by genetic deletion of S1PR1 Matloubian et al showed that among S1P receptors, S1PR1 is indispensable for lymphocyte egress from secondary lymphoid organs and thymus (Matloubian, 2004). FTY720 is phosphorylated by sphingosine kinase (Brinkmann et al, 2002) and the phosphorylated metabolite (FTY720-P) binds to S1PR1 and internalizes it, creating pharmacological S1PR1-null state on lymphocytes, which explains the mechanism of FTY720-induced lymphopenia (Matloubian, 2004).

Figure 4: Metabolic conversion of FTY720 to FTY720-P by sphingosine-kinases. FTY720 is a structural analog of sphingosine. Sphngosine and FTY720 are phosphorylated by sphingosine kinases (Sphk1, 2) to yield S1P and FTY720-P, respectively. FTY720-P represents the biologically active form, and bind to S1PR1. This figure is from Brinkmann et al., 2004.

Figure 5: S1P/S1PR1-dependent T-cell egress from lymph nodes: modulation by FTY720. (A) Naive T cells (Tn) enter LNs via HEVs and emigrate through the sinus-lining endothelium (SLE) of efferent lymphatics in an S1P/S1PR1-dependent manner. When T cells encounter with antigens in the LNs, Tn are activated (Tact) and transiently down-modulate S1PR1; this renders cells unresponsive to the S1P gradient, and, as a consequence, proliferating cells remain in the LNs. At the end of the proliferation phase, Tact up-regulate S1PR1 and egress from LNs in an S1P/S1PR1-dependent step. (B) FTY720, after phosphorylation, acts as ‘super agonist’ at S1PR1 on Tn and Tact, thereby inducing aberrant internalization of S1PR1. As a consequence, Tn and Tact are ‘trapped’ in LNs. Similarly, FTY720 down-modulates S1P1 on thymocytes and B-cells, retaining them in thymus and LN, respectively (not shown). This figure is from Brinkmann et al., 2004.

During immune response, lymphocyte egress is also controlled by S1PR1. Inflammation signals upregulate an early activation marker CD69 on naïve T cells, and upregulated CD69 couples and internalizes S1PR1, which leads to unresponsiveness to S1P (Cyster and Schwab, 2012). This retention allows time for proliferation and acquisition of effector functions. Collectively, S1PR1 is indispensable for lymphocyte recirculation via regulating lymphocyte exit from the thymus and peripheral lymphoid organs.
1.5. Spatial control of S1P generation and receptor activation in vivo

S1P-mediated lympohcyte egress is regulated by gradient of S1P at the tissue-vascular interface (Cyster and Schwab, 2012). Schwab et al showed that S1P is abundant in lymph and blood, but low in lymphoid tissues. Inhibition of S1P-degrading enzyme S1P lyase by treatment with 2-acetyl-4-tetrahydroxybutylimidazole blocked lymphocyte egress, indicating that egress is mediated by S1P gradient that are maintained by S1P lyase activity (Schwab et al, 2005). Beside S1P lyase, LPP3/PPAP2B is also regulates S1P gradient in the thymus, allowing thymocytes to egress (Bréart et al, 2011). LPP3 is expressed in thymic epithelial cells and ECs and conditional deletion of LPP3 either in epithelial cells or ECs inhibits thymocyte egress (Bréart et al, 2011). Source of S1P in the circulation is also well studied. By using Sphk1 and Sphk2-double knockout mice, Pappu et al. showed that blood and lymph S1P is mainly generated by erythrocytes and stromal cells (Pappu et al, 2007). In addition, deletion of Sphk1 and Sphk2 selectively in lymphatic endothelial cells (LECs) (Lyve1-Cre; Sphk1f/fSphk2-/-) leads to a loss of S1P in lymph while maintaining normal in plasma. The mice also showed the defect in lymphocyte egress from LNs and Peyer’s patches (Pham et al, 2010). Further study showed that a S1P transporter Spns2, which transport intracellular S1P outside of cells, is also required for efficient lymphocyte egress (Fukuhara et al, 2012). Spns2 are mainly expressed on ECs and deletion of Spns2 in ECs resulted in defects of lymphocyte egress from the thymus and LNs (Fukuhara et al, 2012). In conclusion, extracellular S1P is produced by ECs and the concentration at tissue-vasucular barrier is tightly controlled by a S1P degrading enzyme. The degradation in tissues makes S1P gradient, acting on lymphocytes to promote their egress from lymphoid organs. On the other hand, recent study showed that using S1PR1-activation reporter mice; S1PR1 is predominantly activated in the vascular structures in lymphoid organs including thymus and LNs rather than in the lymphocytes under the homeostatic condition (Kono et al, 2014).


It has been believed that S1PR1 is expressed on the T cells and play an essential role for T cell egress from efferent lymphatics of the lymphoid tissues. A recent elegant study using S1PR1-activation reporter mice showed that S1PR1 is predominantly activated in the blood and lymph vessels rather than T cells (Kono et al, 2014). This study suggests that lymph-derived S1P acts on lymphatic endothelial cells. However, the role of S1PR1 in the lymphatic endothelial cells is unknown.
Therefore, the aim of this study is to investigate the role of S1P receptor S1PR1 in the lymphatics by developing conditional knockout mice (Lyve1-Cre S1pr1f/f mice) where lymphatic endothelial cells are rendered null for S1pr1.


3.1. Mice

B6;129P2-Lyve1tm1.1(EGFP/cre)Cys/J (Lyve1-CRE), B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J (R26-tdTomato), B6.129S6(FVB)-S1pr1tm2.1Rlp/J (S1pr1f/f), B6.Cg-Tg(CAG-DsRed*MST)1Nagy/J (beta-actin-DsRed) and C57BL6-Tg (CAG-EGFP) 10sb/J (beta-actin-eGFP) were purchased from Jackson Laboratory. Lyve1-CRE/ S1pr1f/f mice were generated by breeding Lyve1-CRE mice with S1pr1f/f mice. Lyve1-CRE mice were bred with Rosa26-tdTomato mice to generate Lyve1-CRE/ R26-tdTomato mice.

3.2. Immunohistochemical analysis

To find lymphatics in the thymus, thymus was sectioned at the thickness of 20 µm. The sections were fixed with methanol at -20°C for 2 minutes and dried for 10 minutes. Sections were washed with 1x PBS three times, and incubated with with 10% normal mouse serum for 30 mins. Sections were further incubated with primary antibodies (Alexa488 conjugated anti Lyve 1, APC conjugated anti CD31, rabbit polyclonal anti-S1PR1 antibodies (H60; Santa Cruz)) over night at 4° C. Samples were washed for three times with 1x PBS. Then they were incubated with secondary antibody (Alexa546 conjugated goat antirabbit IgG) for 1 hour in a dark chamber at room temperature. Sections were washed three times with 1x PBS and mounted with DAPI.

3.3. Flow cytometric analysis

For flow cytometric analysis, cells were incubated with the antibodies {APC-Cy7-conjugated anti-CD4 (GK1.5), PerCP-Cy5.5-conjugated anti-CD8 (53-6.7), Pacific Blue-conjugated anti-B220 (RA3-6B2), BV510-conjugated anti-CD62L (MEL-14), FITC-conjugated anti-CD44 (IM7), PE-Cy7-conjugated anti-CD69 (H1.2F3), Pacific Blue-conjugated anti-HSA (M1/69),Alexa647-conjugated anti-Qa-2 (695H1-9-9), anti-6C10 (SM6C10) (30), V450-conjugated anti-CD45 (30-F11), APC-conjugated anti-CD31 (MEC13.3) or PE-Cy7-conjugated anti-Gp38 (8.1.1)} in PBS supplemented with 2% FCS,1 mM EDTA, 0.1% Sodium Azide and 2% Normal Mouse Serum (FACS media). All of the antibodies were purchased from either from BD Biosciences or BioLegend.

After collecting the thymus and LNs, cell suspension was prepared with 2% FCS in RPMI and transferred to 15 ml tubes, and washed with the same media. Then cells were resuspended with 5 ml of newly-made FACS media. The cell suspension was stained with antibodies including 40 ?g/ml of rat anti-S1PR1 monoclonal antibody (713412; R;D Systems) or isotype control. The cells were washed and incubated with phycoerythrin-conjugated goat anti-rat IgM antibody (Southern Biotech) for 30 minutes on ice. The samples were then washed with FACS media, and filtered and collected into new tubes and kept on ice until flow cytometric analysis. All data were acquired with a LSR Fortessa (BD Biosciences) and analyzed using FlowJo software (FlowJo, LLC).

3.4. Intrathymic injection

CD62L+ CD4+ SP cells were isolated from the thymus using a mouse CD4+ T cell isolation kit (StemCell Tech), followed by positive selection of CD62L+ cells (Miltenyi Biotech). These cells incubated for 30 minutes with either 5 ?M CMTMR or 5 ?M CFSE in RPMI that contained 2% FCS at 37°C. Intrathymic injection was performed as described previouly (Adjali et al., 2005) with some modifications. 1 × 106 cells in 10 ?l PBS were injected into thymus of the recipient mice. Young female mice (less than 4 weeks old) were used as the recipient mice. The single cell suspension directly injected into the thoracic cavity through the skin using a 10-?l Hamilton syringe that was equipped with a 33-gauge needle.

3.5. Adoptive home assays

Lymphocytes were collected from LNs of beta-actin-eGFP mice. 1 × 107 lymphocytes in 200 ?l PBS were intravenously injected into the recipient mice. The recipient mice were then sacrificed 12 hrs after the injection. Donor-derived cells in the inguinal LNs were quantified by flow cytometry.

3.6. Statistical analysis
Differences between groups were evaluated with Student’s t-test for single comparisons or one-way ANOVA, followed by post hoc Tukey tests for multiple comparisons. The statistical analyses were performed using Prism software (GraphPad). A P-value 4) and per group had three mice. Data were analyzed by Student’s t-test and represent the mean ±SD (*P:

Post Author: admin