Lymphangiogenesis, Myeloid Cells and Inflammation

Lianping Xing; Rui-Cheng Ji


Expert Rev Clin Immunol. 2008;4(5):599-613. 

In This Article

Inflammatory Lymphangiogenesis

There is increasing evidence that lymphatic vessels actively participate in acute and chronic inflammation. An increased lymphatic vasculature network has been described in many inflammatory conditions, such as corneal inflammation, wound healing, renal transplant rejection, chronic airway inflammation and inflammatory arthritis, both in human and animal models.[21,47,48,49,50,51] However, the cellular and molecular mechanisms leading to lymphangiogenesis in inflammatory conditions are not well characterized. More importantly, whether lymphangiogenesis plays a beneficial or a detrimental role in inflammation resolution is not clear. The influence of myeloid lineage cells in LEC function at the molecular level is not known, nor is it known if other lymphatic components, such as draining LNs, contribute to inflammatory lymphangiogenesis. Finding answers to these questions will enhance our knowledge of lymphatic biology and provide guidance for development of the lymphatic-based therapies.

In the bone marrow, hematopoietic stem cells give rise to two kinds of multipotent lineage-positive progenitors: common myeloid and lymphoid progenitors, based on their surface markers, gene-expression profile and clonogenic potency.[52] The common myeloid progenitors, dependent on their Epo receptor status, can be further divided into two committed lineage progenitors: megakaryocyte/erythrocyte progenitors and granulocyte/macrophage progenitors. Since granulocyte/macrophage progenitors account for the majority of common myeloid progenitors, they are often referred to as myeloid lineage cells. Common features of myeloid lineage cells are the expression of the CD11b protein on the cell surface of the cells giving rise to macrophages, dendritic cells (DCs) and osteoclasts when they are exposed to different sets of cytokines.[53] Macrophages, DCs and osteoclasts all produce VEGF-C in response to proinflammatory cytokines, indicating the potential importance of myeloid lineage cells in lymphangiogenesis.

Macrophages. Macrophages play an important role in native immunity. Macrophages are generated from a CD11b+ common myeloid precursor. They are secretory cells and produce many factors in response to inflammatory stimuli, including angiogenic and lymphangiogenic growth factors.[54] TNFs and IL-1 stimulate macrophages to produce VEGF-C through the canonical NF-κB signaling pathway.[55]

The involvement of macrophages in inflammation-induced lymphangiogenesis is supported by the following experimental findings:

  • They produce lymphatic factors, such as VEGF-C and PDGF-BB, in response to TNFs;

  • They are present within and around lymphatic vessels in the inflamed area after adaptive transferring into the recipient mice;

  • Most importantly, the elimination of macrophages using clondronate liposomes reduces inflammation-mediated lymphangiogenesis.[21,47,48,56]

Based on these findings, it is proposed that inflammation recruits bone marrow-derived circulating macrophages to the disease sites and produces VEGF-C and other lymphatic factors that stimulate lymphangiogenesis through a paracrine signaling system and/or give rise to LECs directly. The latter claim is supported by observing CD11b+ cells expressing LEC markers and presenting in the lymphatic vessels around inflammation, but not in those without inflammation.[21,56]

The concept of macrophage contribution to lymphangiogenesis raises many open questions. First, it is not known if macrophages contribute to normal lymphatic development or only disease-associated lymphangiogenesis. Second, increased macrophages in inflammation is a well-established phenomenon but the degree of their contribution to lymphatic vessel formation is not clear because relatively few donor-derived macrophages are identified around and/or within the inflammatory lymphatic vessels[21,47,48,56] and their ultimate fate and functional relevance are unknown. Third, it is not clear if lymphatic precursor cells also express some macrophage surface markers, such as CD11b, or if LECs and macrophages share a common progenitor cell. Finally, if macrophages can give rise to LECs in inflammation, one should be able to induce macrophage-like cells to express LEC markers invitro. However, while macrophages isolated from the peritoneal cavity, after they are activated by thioglycolate injection or from inflamed joints of TNF-Tg mice, express high levels of VEGFR-3, podoplanin and LYVE-1,[21,55] VEGF-C treatment does not induce LEC marker gene expression in primary bone marrow-derived macrophages,[21] suggesting that VEGF-C alone cannot stimulate macrophages to express LEC markers invitro. We found that TNFs, IL-1 and RANKL, had no effect on mRNA expression of LEC markers using bone marrow-derived macrophages (ZhangQ, Guo R, Boyce BF, Xing L, Unpublished Data). A recent study found that IL-1 stimulates bone marrow macrophages to express LEC markers and form tube-like structures in Matrigel™, only if they are cultured in L929 cell-conditioned medium.[57] The L929 cell line is derived from mouse connective tissue and the factors from the L929 cell-conditioned medium responsible for macrophage-LEC transition are not known. Nevertheless, this report indicates that, under certain conditions, macrophages do have the potential to behave like LECs.

It is clear, at least in mice, that lymphatic vessels develop during embryonic development by sprouting and lymphatic specialization of endothelial cells located on one side of embryonic cardinal veins.[58,59] The commitment of common vascular precursor cells to LECs is determined by Prox-1. VEGF-C and other lymphatic growth factors function at the stage thereafter. However, it is not known where the LEC precursors come from in the inflammation-induced lymphangiogenesis in adult mice. Bone marrow-derived CD11b+ myeloid cells play an important role in postnatal angiogenesis through the interaction between SDF-1/CXCR4 and the VEGF-A signaling pathway;[60] whether or not this is also the case in lymphangiogenesis needs to be explored.

Dendritic Cells. Dendritic cells are antigen-presenting cells that are critical for the induction and regulation of adaptive immune responses.[61] Inflammation or tissue injury stimulate the release of cytokines or cytokine-like molecules, such as high mobility group box 1[62] and purine nucleotides,[63,64] to the extracellular space and trigger DC maturation and/or migration to lymphoid organs. Under normal physiological conditions, a small number of DCs constitutively emigrate from tissues and enter regional draining LNs via the afferent lymphatics,[65] a process that is greatly accelerated by inflammation.[66,67] One of the mechanisms is that the inflammatory signal promotes DC maturation and CCR7 expression, the chemokine receptor required for DC trafficking to LNs.[68,69] Deficiency of CCR7 or its ligands, CCL19 and CCL21, leads to impaired DC migration into draining LNs and abnormal LN architecture.[70]

Dendritic cells and macrophages are derived from a common myeloid progenitor-expressing CD11b protein but they have different migratory capacities. The number of macrophages in the airways is 100-fold more than the number of DCs but many more DCs than macrophages migrate from the airways to lung-draining LNs,[71,72] underlining a critical difference in the distinct immune-priming capacity of these two cell types.

Many studies demonstrate that DCs in the inflammatory areas express VEGF-C proteins by immunostaining with anti-VEGF-C and DC surface markers. Similar to macrophages, TNFs, IL-1 and other pro-inflammatory cytokines are proposed to be stimuli for DCs to express VEGF-C. However, whether DCs release VEGF-C to affect LEC function indirectly or DCs themselves can give rise to LECs, to directly contribute to lymphangiogenesis, is not known. One distinguished difference between DCs and macrophages is that DCs express the receptor activator of nuclear factor κB (RANK), which is activated by the RANK ligand (RANKL). RANKL/RANK signaling affects DC function and survival.[73,74] We found that RANKL treatment strongly enhances VEGF-C expression in osteoclasts, which also express RANK.[75] It will be interesting to determine if RANKL stimulates DCs to produce VEGF-C in normal and disease conditions.

Osteoclasts and Precursors. Osteoclasts are bone-resorbing cells that play important roles in normal and pathologic bone remodeling. Osteoclasts are derived from CD11b-expressing myeloid precursors.[76,77] Osteoclast formation requires the expression of macrophage-colony stimulating factor and RANKL by accessory cells. The RANKL/RANK interaction in osteoclast precursors triggers a cascade of signal transducing events and sequentially activates the transcription factors, NF-κB, AP-1 and NFATc1, leading to the differentiation of osteoclast precursors to osteoclasts.[78,79,80] Recent studies demonstrate that osteoclasts are involved in more complex processes than simply resorption of bone. Osteoclasts can be activated to secrete factors that act through autocrine and paracrine mechanisms to contribute to inflammation and autoimmunity.[81,82] In addition, PDGF-BB has been detected in the conditioned medium of mouse macrophage/osteoclast precursor Raw 264.7 cells treated with RANKL.[83] These findings indicate that osteoclasts also function as immunomodulators and affect the functions of other processes in addition to bone resorption.[84,85]

We found that osteoclasts produce a large quantity of VEGF-C (more than tenfold the normal level) in response to RANKL treatment. RANKL-induced VEGF-C expression was abolished in cells from NF-κB p50 and p52 double-knockout mice. This indicates that VEGF-C is a novel target of RANKL signaling in osteoclasts, which is mediated by NF-κB signaling.[75] Using LYVE-1 and podoplanin antibodies, we found that there are no lymphatic channels in the bone marrow cavities of mice. In contrast to these findings within bone, we found remarkably enlarged lymphatic vascular networks in the soft tissue around the inflamed joints of TNF-Tg arthritic mice,[55] where RANKL levels are high[86,87] and osteoclasts strongly express VEGF-C. Thus, osteoclast-derived VEGF-C may play a more important role in pathologic conditions where RANKL and inflammatory cytokines are highly expressed, such as in arthritic joints and in certain types of cancer.[88,89] It remains to be determined if osteoclasts play a significant role in lymphatic vessel formation during development but the absence of osteoclasts and LNs in mice deficient in RANKL, RANK or NF-κB p50/p52[4,35] suggests that this merits further study.

The lymphatic system drains lymph and transfers DCs from peripheral tissues to local LNs to initiate and orchestrate immune responses.[65] The CCR7 and its ligands, CCL19 and CCL21, modulate the migration of DCs through multiple signaling pathways within the LNs.[90] Lymph also contains hematopoietic stem and progenitor cells.[91] These cells originate in the bone marrow, enter the blood, and traffic to peripheral organs, where they reside for at least 36h before entering draining lymphatics to return to the blood and, eventually, the bone marrow. Within the extramedullary tissues of peripheral organs, these cells proliferate and give rise to tissue-resident myeloid cells, preferentially DCs. Sphingosine-1-phosphate receptors mediate the egress of hematopoietic stem and progenitor cells from extramedullary tissues into the lymph. Thus, hematopoietic stem and progenitor cells may promote the local production of tissue-resident innate immune cells both under steady-state conditionsand in response to inflammatory signals.[91]

The immunization of mouse paws with the antigen keyhole limpet hemocyanin emulsified in complete Freund's adjuvant leads to a strong LN response and expansion, resulting in LN lymphangiogenesis and the migration of DCs to the nodes. Increased LN lymphangiogenesis is prevented by the systemic administration of VEGFR-2-neutralizing antibody, suggesting a VEGF-A-mediated mechanism. B cells within the LNs are a primary source of VEGF-A and the deletion of B cells abolishes immunization-induced lymphangiogenesis.[92]

In delayed-type hypersensitivity reaction-induced ear inflammation, the number of LECs in inflamed ears and draining LNs are significantly increased. Increased lymphangiogenesis is blocked by systemic administration of a VEGF-A inhibitor. Interestingly, only the VEGF-A protein, but not VEGF-A mRNA, is increased in the LNs, while both VEGF-A protein and mRNA are elevated in the ear, raising a hypothesis that VEGF-A is produced at the distant inflammatory areas and transported to the LNs through the lymphatic pathway. Thus, LN lymphangiogenesis can be controlled remotely by lymphangiogenic factors produced at the site of inflammation.[93] Furthermore, the deletion of B cells does not affect the LN lymphangiogenesis in this model.[93]

Using TNF-Tg mice as a model of chronic inflammatory arthritis and serum-induced arthritis as a model of acute inflammatory arthritis, we assessed the changes in volume of draining PLN and synovial volume over time with MRI.[94] The PLN volume is remarkably increased in TNF-Tg mice with established arthritis. The PLNs of TNF-Tg mice have increased LYVE-1+ lymphatic sinuses, with dilated lumens often containing clusters of cells.[94] In serum-induced arthritis, the PLN volume is also increased but the LYVE-1+ sinuses are much narrower and no cells are present in the lumen. The functional consequences of different LN phenotypes are not currently clear. In preliminary studies, the VEGFR-3-neutralizing antibody reduces the size of PLNs in both models but causes more joint lesions, which are only seen in TNF-Tg mice. These findings suggest that, although the size of draining LNs is remarkably increased, they must play a different pathological role in chronic versus acute inflammatory arthritis, which needs to be investigated in detail.

Although larger draining LNs have been observed in patients with rheumatoid arthritis (RA) since the 1950s[95,96] and the MR images of PLNs have been suggested to serve as simple and useful markers in differentiating RA from osteoarthritis (OA) patients,[97,98] the precise role of LN lymphangiogenesis and LN drainage in inflammation development and resolution is still not clear. In addition, the ultrastructural changes in the lymphatic sinuses within LNs have not been studied under inflammation. This is important because inflammatory stimuli may directly affect LEC structure and function, as well as damaging surrounding matrix components, which would further promote lymphatic dysfunction. Recently, we applied the 5'-nucleotidase (5'-Nase) cerium staining in PLNs from wild-type BALB/c mice using a similar protocol that we have developed to identify lymphatic vessels in the skin[99] and other tissues.[100] 5'-Nase cerium staining is an enzyme-histochemical staining method that specifically identifies the LECs ultrastructurally. We observed that 5'-Nase-positive lymphatic sinuses localize near the high endothelial venue and they contain numerous lymphocytes, suggesting that some lymphocytes may migrate into the lumen of these lymphatic sinuses (Figure 2). How these lymphocytes function within the LNs and whether they interact with antigen-presenting DCs transporting antigens via intranodal lymphatic sinuses are open questions, which can be investigated using the newly developed multiphoton microscopy.[101]

Figure 2.

Electron microscopy of lymph node lymphatic sinuses identified by 5'-Nase cerium staining. The popliteal lymph nodes of wild-type BALB/c mice were stained by 5'-Nase cerium. The picture shows 5'-Nase-positive LSs near the HEV. Numerous lymphocytes are present in the lumen of these LSs. The insert is an enlargement of a position of the LSs showing the cerium participates (arrows) mainly in the luminal surfaces of the lymphatic endothelial cells. 5'-Nase = 5'-nucleotidase; HEV = High endothelial venue; LS = Lymphatic sinus.

It is conceivable that in the case of inflammation, lymph drained from the inflammation sites contains cytokines and inflammatory cells. When these factors reach the draining LNs, they will affect LN lymphangiogenesis and function, which may influence the pathology of the primary inflammation. Newly developed invivo imaging technologies will help to solve these questions by assessing lymphatic draining function longitudinally in various mouse models of inflammation.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: