Angiogenesis or the lack thereof is a key event in the development and progression of major pathological conditions including diabetic retinopathy, psoriasis, rheumatoid arthritis, cardiovascular diseases, and tumor growth (reviewed by Carmeliet and Jain 2000). Much of the interest in angiogenesis research is driven by the idea of therapeutic intervention. Pro-angiogenic proteins like VEGF and VEGF-C have been used in clinical studies to boost the growth and perfusion of blood vessels after vascular injuries like myocardial infarction (Isner et al. 1996; Witzenbichler et al. 1998). Also, because of convincing evidence that solid tumors are angiogenesis-dependent, anti-angiogenic compounds have entered clinical trials (Hanahan and Folkman 1996). Similar to developing embryos, tumors can only grow beyond the limits of diffusion by establishing a vascular network for the distribution of oxygen and nutrients. An imbalance between pro-angiogenic and anti-angiogenic factors results in vessel ingrowth and is followed by rapid tumor expansion (Hanahan and Folkman 1996). Despite encouraging results obtained in several mouse models of anti-angiogenic tumor therapy and pro-angiogenic therapy of ischemia, success in preclinical and clinical trials is limited (Cao 2001; Hammond and McKirnan 2001). Caution has been recommended (Carmeliet 2000b) and also principal objections against the safety and feasibility of such therapies have been raised (Blagosklonny 2001). In developed countries the angiogenesis-related cardiovascular and neoplastic diseases are major health issues, whereas lymphatic disorders are relatively rare.
Insufficiency of lymphatic transport can result in lymphedema, which can either be hereditary or with unknown etiology (primary lymphedema), or a consequence of a previous disease or trauma (secondary or acquired lymphedema). Iatrogenic lymphedema, and especially postmastectomy edema, represents probably the most common lymphatic condition in civilized countries. Its incidence has been estimated to between 6 to 30% after surgical treatment of mammary carcinoma (Petrek and Heelan 1998). Surgical damage to the lymphatics is commonly thought to be the cause of post-surgical edema, but also other reasons such as compromised venous return seem to be involved (reviewed by Pain and Purushotham 2000). The worldwide most common cause of lymphedema is, however, filariasis - mostly caused by infection with Wuchereria bancrofti or Brugia malayi. These parasitic nematodes are transmitted by mosquito bites. The parasite lives and reproduces in the lymphatic system causing a massive lymphatic dilatation in early stages of the disease. In in advanced disease, lymphatic transport is blocked leading to an extreme enlargement of the limbs or other areas of the body called elephantiasis (Rao et al. 1996; Dreyer et al. 2000).
In hereditary lymphedema lymphatic vessels can either be hypoplastic or hyperplastic, but non-functional. In addition to some broad-spectrum syndromes such as Ulrich-Turner and Noonan, that are associated with lymphedema, a large number of distinct lymphedema syndromes has been described. The current phenotypic classification seems inadequate, based on recent clinical and genetic data showing that the same genetic cause can give rise to several distinct phenotypes, and the same phenotype can be caused by distinct genetic alterations (Kääriäinen 1984; Ferrell et al. 1998; Finegold et al. 2001).
Type I hereditary lymphedema (Milroy disease, OMIM 153100) is an early onset form of hereditary lymphedema. In these patients, the initial superficial lymphatics of edematous areas cannot be demonstrated by fluorescence microlymphography and are believed to be absent or highly hypoplastic. However, in non-edematous areas superficial lymphatics are present (Bollinger et al. 1983).
In some families inheritance is strongly linked to dominant missense mutations in the VEGFR-3 gene on chromosome 5 (Karkkainen et al. 2000). However, penetrance is incomplete or variable in these families. Other families showed additional linkage to multiple loci on chromosomes 3, 11 and 18, unrelated to any known target genes. Thus, oligogenic pattern of inheritance, modifier genes and environmental factors might be necessary to explain the hereditary patterns. Type II hereditary lymphedema (Meige disease, OMIM 153200) differs from type I by a later disease onset (around puberty), and its etiology appears even more complex with only 10% of the families showing dominant pattern of inheritance and a penetrance of 40% (Holberg et al. 2001).
Dominant mutations in the transcription factor FOXC2 have been identified as the cause of lymphedema-distichiasis syndrome (OMIM 153400; Fang et al. 2000; Finegold et al. 2001). In addition to lymphedema distichiasis three other lymphedema syndromes co-segregated with FOXC2 mutations: type II hereditary lymphedema, lymphedema/ptosis (OMIM 153000) and lymphedema/yellow-nail syndrome (OMIM 153300). All indentified FOXC2 mutations result in a truncated protein and the observed phenotype is likely a result of haploinsufficiency.
Lymphangiectasia is a lymphatic disorder characterized by dilated dysfunctional lymphatics. The condition can be limited to a specific organ. The lungs are affected in hereditary pulmonary cystic lymphangiectasia (OMIM 265300) and the intestine in Hennekam lymphangiectasia-lymphedema syndrome (OMIM 235510; Hennekam et al. 1989; Gilewski et al. 1996). The causes of lymphangiectasia and lymphedema are thought to be similar and both conditions do occur jointly in syndromes such as Noonan Type I (OMIM 163950), Hennekam or hereditary intestinal lymphangiectasia (OMIM 152800). Similar to lymphedema, also acquired forms are known (Celis et al. 1999).
Occasionally, neoplasms are derived from lymphatic endothelial cells. Lymphangiomas constitute approximately 5% of all benign lesions of infancy and childhood (Zadvinskis et al. 1992). Since lymphangiomas can present either as localized mass or as a diffuse tumor, it is questionable whether all lymphangiomas represent true neoplasms (Scalzetti et al. 1991). Unlike lymphangioma, lymphangiosarcoma is a true malignant lesion of lymphatic endothelial cells. Mostly it occurs as a complication of post-mastectomy edema (Stewart Treves syndrome; Janse et al. 1995).
The lymphatic system serves as the primary pathway for metastatic spread of tumor cells to regional lymph nodes, and possibly also to distant organs. The prognostic value of lymph node metastasis was recognized long before the concept of lymphangiogenesis both within and adjacent to tumors became widely accepted (Fisher et al. 1969; Carter et al. 1989). Tumor lymphangiogenesis occurs in experimental models in mice with important implications for metastasis (Karpanen et al. 2001; Mandriota et al. 2001; Skobe et al. 2001; Stacker et al. 2001). It is, nevertheless, still unclear to what extent lymphangiogenesis occurs in human cancers, and what the consequences are for cell dissemination (Leu et al. 2000; Birner et al. 2001; Jackson et al. 2001; Schoppmann et al. 2001).
The regeneration of lymphatic vessels was observed by Clark and Clark already in 1932. However, expertise in therapeutic lymphangiogenesis is just emerging (Karkkainen et al. 2001). Several recent papers have been shown that in angiogenic therapy the balance between harm and help is not trivial (Masaki et al. 2002) and the use of single molecules is likely to be insufficient (Richardson et al. 2001). Influencing the molecular decision makers such as Hypoxia-inducible factor-1 (HIF-1) instead of the effector molecules such as VEGF might be easier than the futile attempt to mimic the temporally and spatially complex growth factor cocktail of mother nature (Vincent et al. 2000; Elson et al. 2001). Whichever pro-angiogenic therapy, it is possible that pro-angiogenic or pro-lymphangiogenic therapy leads to accelerated cancer progression and metastasis.
So far no good animal model has been developed for aquired lymphedema, but a mouse model of hereditary lymphedema does exist: the Chy mouse. Chy mice carry a mutant Vegfr-3 allel coding for a kinase-dead receptor. In this model VEGF-C was used to compensate for insufficient VEGFR-3 signaling. This resulted in the growth of new functional lymphatics (Karkkainen et al. 2001). It is not clear why VEGF-C does induce lymphatic sprouting in these mice, but fails to do so in the keratin-14 VEGF-C transgenic mice. The VEGF-C transgenic mice were edematous, indicating that also too much VEGFR-3 signaling can impair lymphatic function (I). From the visceral lymphatics in the Chy mouse, only the lacteals are aplastic although the VEGFR-3 mutation is assumed to be dominant negative in all lymphatic endothelial cells and there is circumstantial evidence that lymphangiogenesis might occur independently of VEGFR-3 activation (Taija Makinen, personal communication). Thus many molecular players are still to be identified. In conclusion, therapeutic modulation of vascular growth in pathologic conditions represents the major challenge in the fields of both angiogenesis and lymphangiogenesis.