Of mice and men; the translational physiology of a genetic form of lymphoedema

The lymphatic system was for a long period the Cinderella of physiology, exciting little research interest. In the 1990s, however, a defining stimulus to lymphatic research came with the discovery of the vascular endothelial growth factor (VEGF) family of proteins. The receptor VEGFR-3 and its ligand, VEGF-C (and, recognized later, VEGF-D), were soon identified as the first signals to act specifically on the lymphatic system and be responsible for lymphangiogenesis (Kaipainen et al. 1995; Joukov et al. 1996). Further progress at the molecular level was the recognition of the first marker for lymphatic endothelial cells, namely LYVE-1 (Jackson, 2004). Lymphatic biology has now become a hot topic in science and medicine, with the lymphatic system seen as an important player in major disease processes such as cancer spread, infection and inflammation, asthma, transplant rejection and of course lymphoedema.

Developments in molecular biology have now revealed a whole raft of genes and proteins that act specifically or mainly on the lymphatic system (Alitalo et al. 2005). Consequently, a major current task for physiologists is translation, to determine the significance of the molecular findings in vitro for intact tissues and whole body physiology in vivo. In this issue of The Journal of Physiology, Karlsen et al. (2006) make an important contribution to translational lymphatic physiology by investigating the physiological consequences of disrupting the lymphangiogenesis receptor, VEGFR-3, and hence lymphatic development, in genetically engineered mice. Besides its physiological interest, this is of great relevance to understanding a human genetic disease, Milroy disease. Milroy disease is a form of hereditary lymphoedema of the lower legs caused by inactivating mutation of the Vegfr3 gene, resulting in hypoplasia of the lymphatic capillary network.

In the mouse model, lymphatic capillaries appeared to be absent from skin biopsies from the forepaw, hindpaw, thigh and back, but only the fore- and hindpaws had visible oedema. The physiological consequences of the lymphatic aplasia were assessed by measuring interstitial fluid volume (IFV), interstitial fluid pressure (P_if) and interstitial colloid osmotic pressure (COP_if) in the various regions, before or after an intravenous fluid load of 15% body weight. P_if and IFV were significantly increased only in the visibly swollen tissues (fore- and hindpaw), whereas COP_if was increased in all skin areas studied. Particularly striking was the response to a volume load, namely a fourfold increase in IFV and significantly greater increase in P_if than in wild-type mice. Measurements of tissue cytokines provided no support for the traditional hypothesis that the raised protein levels in lymphoedema trigger an inflammatory response. The findings thus demonstrate the key role of Vegfr3 and lymphatic capillaries in tissue fluid homeostasis, and especially their importance in clearing an acute increase in blood capillary filtrate.

Although there are striking similarities between the mouse Vegfr3-mutated lymphoedema model and human Milroy disease (Brice et al. 2005), there are also intriguing differences. Oedema developed at birth in the mice and had been present for only 3–4 months at the time of study, whereas Milroy lymphoedema (likewise presenting at birth) has been present for years when investigated. Moreover, human Milroy patients have moderate lower limb swelling but no upper limb swelling, contrasting with both fore- and hindpaw swelling in mice. The latter developed despite the far smaller gravitational enhancement of capillary filtration pressure in mice than men. The difference may arise in part from the reported total aplasia of the dermal lymphatic capillary network in Vegfr3-mutated mice; in human Milroy disease dermal lymphatic vessels can be seen on biopsy but appear to be non-functional in the feet, though functional in the arms, as assessed by fluorescence microlymphography (Mellor et al. 2005). Intriguingly, Karlsen et al. found lymphatics in the subcutis despite the dermal lymphatic aplasia, indicating that the dependence of lymphatics on Vegfr3 may be tissue specific.

The findings of Wiig's group support a recent shift in emphasis to the importance of lymph transport rather than venous capillary reabsorption for interstitial fluid homeostasis (Levick & Mortimer, 1999). Traditionally it has been taught that the arterial ends of capillaries filter fluid while the venous ends reabsorb most of this filtrate. Modern evidence indicates, however, that in most vascular beds there is a net filtration along the entire length of well-perfused capillaries, albeit dwindling to almost zero at the venous end. Although venous capillary pressure falls below plasma COP, the substantial COP_if and P_i tip the balance of the Starling forces into a slight filtration force, even in venous capillaries. Values for COP_if and P_i reported by Wiig's group in wild-type mice reinforce this view. Thus the major responsibility for the drainage of interstitial fluid lies with the lymphatic system, rather than sustained fluid absorption by venous capillaries. The paper by Karlsen et al. lends further weight to the crucial role of the lymphatic to tissue volume homeostasis, and highlights intriguing differences as well as similarities between the mouse model and human Milroy disease.