Interaction of B and T lymphocyte subsets with high endothelial venules in the rat: binding in vitro does not reflect homing in vivo.
Lymphocytes continuously migrate through the body, and their efficient extravasation from the blood via high endothelial venules (HEV) is essential for initiating an appropriate immune response. Most investigations have focused on the lymphocyte/HEV interaction in vitro. However, to what extent such systems reflect the situation in vivo is not known. It is also unclear whether lymphocyte subsets immigrate into the HEV in proportion to their presence in the blood, and whether import capacity is limited by the HEV. When rat mesenteric lymph node lymphocytes were incubated in vitro on cryostat sections, the well-known preferential binding of B lymphocytes to HEV of Peyer's patches (PP) and T cells to HEV of axillary lymph nodes (axLN) was observed (axLN vs. PP: B lymphocytes 21.2 +/- 5.0% vs. 40.6 +/- 11.0%, T lymphocytes 84.6 +/- 6.3% vs. 56.5 +/- 12.9%). However, when labeled mesenteric lymph node lymphocytes were injected and their location within the HEV was analyzed 15 min later, no preferential interaction was seen. After injection of labeled thoracic duct lymphocytes, the percentage of labeled cells among B and T lymphocytes in the blood was significantly different (4.4 +/- 0.9% vs. 8.9 +/- 3.6%), whereas that in HEV of axLN (19.0 +/- 6.4% vs. 16.6 +/- 6.0%) and PP (30.6 +/- 6.1% vs. 33.9 +/- 4.4%) was comparable. Although the number of injected lymphocytes was similar in magnitude to the total blood lymphocyte pool, after injection there was no increase in lymphocyte numbers in the HEV. Thus, the adhesion assay in vitro does not completely reflect immigration into HEV in vivo. In addition, our data suggest that both the availability of lymphocyte subsets in small venules and the immigration rate into HEV are actively regulated in vivo.