Inflammation-induced lymph node lymphangiogenesis is reversible

Abstract

The extent of lymph node metastasis is a prognostic indicator of disease progression in many malignancies. Current noninvasive imaging technologies for the clinical assessment of lymph node metastases are based on the detection of cancer cells and commonly suffer from a lack of sensitivity. Recent evidence has indicated that the expansion of lymphatic networks (ie, lymphangiogenesis) within tumor-draining lymph nodes might be the earliest sign of metastasis. Therefore, we recently developed a noninvasive imaging method to visualize lymph node lymphangiogenesis in mice using radiolabeled antibodies against the lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1) as well as positron emission tomography (PET). This technique, termed anti-LYVE-1 immuno-PET, was found to be very sensitive in the detection of metastasis to the lymph nodes. However, lymphatic vessel expansion to the lymph nodes can also be induced by inflammation, and it is currently unclear whether such vessel expansion is reversed once inflammation has resolved. Detection of residual inflammation-induced lymph node lymphangiogenesis, thus, might hamper the identification of metastasized lymph nodes. In this study, we therefore used a well-established mouse model of inflammation in the skin to investigate whether lymphatic vessels in the lymph nodes regress on resolution of inflammation. Our data reveal that the lymphatic network indeed regresses on the resolution of inflammation and that we can image this process by anti-LYVE-1 immuno-PET.

Figure 1

Inflammation-induced reversible increase of ear thickness. A: The ears of wild-type FVB mice were repeatedly treated with oxazolone to induce inflammation in the skin, which resulted in increased ear thickness (inflammation group). After the oxazolone treatment was discontinued, the ear thickness gradually decreased (regression group) (n = 3 mice per group). Arrow: Time point of positron emission tomographic (PET) imaging of the inflammation group. Arrowhead: Time point of PET imaging of the regression group. B, C: Representative pictures of ears of mice from the control (B), inflammation (C), and regression group (D) at the day of PET imaging. The redness and swelling induced by the delayed-type hypersensitivity (DTH) response (panel C) had subsided in the regression group (panel D). E: The lymph node weights increased on inflammation and decreased again after inflammation resolution. F: Interferon (IFN)-γ mRNA levels were increased in the inflamed compared to the control and regressed lymph nodes. Data represent mean ± SD * P < 0.05. G–K: mRNA levels of the lymphangiogenic factors VEGF-A (G), VEGF-C (H), VEGF-D (I), and hepatocyte growth factor (HGF) (J), and of the lymphatic transcription factor Prox1 (K) were not significantly changed in control, inflamed, and regressed lymph nodes.

Figure 2

In vivo positron emission tomographic (PET) imaging of the lymphatic vessel regression in superficial parotid lymph nodes using ¹²⁴I-anti-lymphatic vessel endothelial hyaluronan receptor-1 (LYVE-1) antibody. A–C: Normalized serial PET sections of mice from the control (A), inflammation (B), and regression groups (C) (n = 3 per group) with higher magnifications of the cervical regions below. The accumulation of ¹²⁴I-anti-LYVE-1 antibody in the superficial parotid lymph nodes (arrows) was stronger in inflammation than in the control group. Regression of inflammation subsequently resulted in reduced accumulation of ¹²⁴I-anti-LYVE-1 antibody in the lymph nodes.

Figure 3

Reversible expansion of the lymphatic network in the lymph nodes on inflammation. A: Radioactivity in the superficial parotid lymph nodes of the mice was measured subsequent to positron emission tomographic (PET) imaging. Inflammation in the ear skin resulted in increased antibody accumulation in the draining lymph nodes. On inflammation resolution, antibody accumulation was decreased again (n = 3 mice per group; cpm = counts of gamma quantum stemming from ¹²⁴I decays per minute). B–D: Immunofluorescence analysis of a representative tissue section from an inflamed superficial parotid lymph node. The injected ¹²⁴I-anti-LYVE-1 antibody (B) completely co-localized with LYVE-1 stained lymphatic vessels (C); (merge, D). Scale bars: 100 μm. E–G: Higher magnifications of panels B–D, respectively. Scale bars: 50 μm. H–J: Microradiographs of tissue sections of control (H), inflammation (I), and regression groups (J) mice following immuno-PET imaging reveal increased accumulation of the injected ¹²⁴I-anti-LYVE-1 antibody (black signal) in the inflamed compared to the regressed or the control lymph node. Scale bars = 100 μm. K–M: Low magnifications of micrographs of control, inflamed, and regressed lymph nodes, respectively. Scale bars: 200 μm.