London, United Kingdom

Correspondence: Professor Adrian S. Woolf, Room 219, Nephro-Urology Unit, Institute of Child Health, University College London, 30 Guilford Street, London WC1N 1EH, UK. E-mail: a.woolf@ich.ucl.ac.uk

The podocyte is an in vivo source of vascular endothelial growth factor (VEGF), and VEGF-A maintains the morphologic integrity of the normal maturing glomerulus¹. More controversial is whether renal tubules express VEGF, and whether, here, the factor has roles in health or disease. Do renal tubules express VEGF in vivo? VEGF-A can be immunodetected in proximal tubules and distal nephron, where protein levels and distribution can change markedly after acute tubular necrosis and subtotal nephrectomy^2,3,4. In these contexts, VEGF-A protein levels correlate with peritubular capillary behavior, with either angiogenesis or capillary loss being observed, depending on the type of insult (e.g., toxic, ischemic, remnant kidney), the time after insult, and the experimental animal species (e.g., rat or mouse) and strain^2,3,4. Kang et al³ noted that, in several models of renal disease, the integrity of peritubular capillaries has a positive correlation with renal function; furthermore, in the rat remnant kidney, tubular VEGF-A becomes diminished, and the administration of VEGF-A minimizes loss of peritubular capillaries and stabilizes renal function³. Conversely, in mouse remnant kidneys, the progression of renal tubular lesions correlates with a robust angiogenic response by the peritubular microcirculation associated with an overall increase of renal VEGF-A protein⁴.

The above data support a functional relationship between tubular VEGF-A and the state of the adjacent renal cortical microcirculation. A skeptic might, however, argue that the finding of immunoreactive VEGF-A in tubules simply reflects the fact that these epithelia take up the factor from either the tubule lumen or the microcirculation; in other words, the data do not prove that tubules make VEGF-A in vivo. On the other hand, there are several in situ hybridization studies that support the idea that renal tubules express VEGF-A transcripts in health, and that these levels can change after physiologic stress and in disease states. For example, Simon et al⁵ reported that human fetal and adult kidney collecting ducts expressed VEGF transcripts, Marti and Risau⁶ noted that adult mice expressed the factor in medullary ducts, and that here, VEGF was up-regulated with systemic hypoxia, and Yuan et al² found that proximal tubules of adult mice express VEGF-A transcripts and protein, and that both were down-regulated after folic acid–induced acute tubular necrosis.

To further define factors that modulate the expression of VEGF-A by renal tubules, investigators have turned to studies of cultured cells. Kidney tubule cells express VEGF-A when grown in vitro, and levels of mRNA and protein can be up-regulated by hypoxia and other stimuli, such as transforming growth factor 1, and down-regulated by interleukins^3,7,8. Binding of active TGF to cell surface receptor serine/threonine kinases leads to a series of biochemical reactions whereby the signal is transduced to the cell nucleus; this involves the direct phosphorylation of receptor-mediated Smad 1, 2, 3, 5, and 8, intracellular proteins which are expressed in most cells⁹. The expression of Smad 6 and 7 proteins are more tightly regulated, and these inhibitory Smads are induced by TGF in an autoregulatory feedback system⁹.

In a study reported in this issue of Kidney International, Nakagawa et al¹⁰ show that the up-regulation of VEGF-A expression by rat proximal tubule cells exposed to TGF1 is linked to the Smad system, and the authors have additionally begun to explore the complexity of TGF1 actions with regard to the responses of other molecules, which themselves modify angiogenesis. These include: (1) the soluble form of Flt-1 (also called VEGF receptor type 1), a molecule which binds to VEGF-A but which does not transduce an angiogenic response; Flt-1, therefore, acts as a natural antagonist of VEGF-A¹¹; and (2) thrombospondin-1 (TSP-1), a molecule which activates latent TGF-1, and which is itself anti-angiogenic⁴. Nakagawa et al¹⁰ found that TGF1 increased expression of both VEGF-A and TSP-1 mRNA and protein in proximal tubule cells, and also increased Smad 2 and 3 phophorylation/activation. Furthermore, forced overexpression of Smad7 inhibited all of these TGF1-induced effects. Using cells from mice which were null-mutant for specific Smad genes, the authors went on to demonstrate that Smad3 was essential for the VEGF-A response, while TSP-1 and soluble Flt-1 stimulation required Smad2; a potential criticism of this part of the study is that fibroblasts, rather than tubular cells, were used. In the last part of the study, the authors investigated the effects of conditioned media on proliferation of human umbilical venous endothelial cells as a surrogate marker of "angiogenesis." Media from TGF1-stimulated Smad2 null-mutant fibroblasts enhanced proliferation, while media conditioned by Smad3 null-mutant fibroblasts failed to do so. This was consistent with above observations that the balance of expressed angiogenic/antiangiogenic factors presumably differs between wild-type, Smad2, and Smad3 mutant cells; similar experiments were not, however, reported with null-mutant proximal tubule cells.

So, do kidney tubules serve an angiogenic soup? As outlined above, there is plenty of evidence that they express VEGF-A in vivo, where levels correlate with expansion or regression of peritubular capillaries. Final proof for the postulated functional roles of tubule-derived VEGF-A in health and disease will require the generation and study of genetically engineered mice in which VEGF-A is ablated, specifically in parts of the renal tubule (i.e., an inducible Cre-lox system) whereby VEGF-A is deleted in proximal tubules or collecting ducts, perhaps using -glutamyl transpepidase and aquaporin 2 promoters to drive Cre recombinase expression. Similar strategies, with VEGF-A ablated in podocytes, were informative for understanding glomerular biology¹. In addition, the work of Nakagawa et al¹⁰ alerts us to the fact that kidney tubule cells in vitro, and therefore, probably in vivo, also express a range of antiangiogenic factors; hence, the biological response of an endothelial cells near a tubule will depend on the balance of epithelial-derived stimulatory and inhibitory factors. The fact that specific intracellular molecules (e.g., Smad2 and 3) affect this balance suggests a possibility that endothelial responses could be therapeutically shifted towards angiogenesis (e.g., in states such as kidney tubular atrophy associated with capillary loss), or antiangiogenesis (e.g., in renal cancers where vessels maintain tumor viability), if the activation of specific Smad proteins could be chemically modulated.

Top of page

References

References

1.	EREMINA V & QUAGGIN SE. The role of VEGF-A in glomerular development and function. Curr Opin Nephrol Hypertens 2004; 13: 9−15.
2.	YUAN HT, LI XZ & PITERA JE et al. Peritubular capillary loss after mouse acute nephrotoxicity correlates with downregulation of vascular endothelial growth factor-A and hypoxia-inducible factor-1. Am J Pathol 2003; 163: 2289−2301.
3.	KANG D-H, KANELLIS J & HUGO C et al. Role of the microvascular endothelium in progressive renal disease. J Am Soc Nephrol 2002; 13: 806−816.
4.	PILLEBOUT E, BURTIN M & YUAN HT et al. Proliferation and remodeling of the peritubular microcirculation after nephron reduction. Association with the progression of renal lesions. Am J Pathol 2001; 159: 547−560.
5.	SIMON M, GRONE HJ & JOHREN O et al. Expression of vascular endothelial growth factor and its receptors in human renal ontogenesis and in adult kidney. Am J Physiol 1995; 268: F240−F250.
6.	MARTI HH & RISAU W. Systemic hypoxia changes the organ-specific distribution of vascular endothelial growth factor and its receptors. Proc Natl Acad Sci USA 1998; 95: 15809−15814.
7.	KIM B-S, CHEN J & WEINSTEIN T et al. VEGF expression in hypoxia and hyperglycaemia: Reciprocal effect on branching angiogenesis in epithelial-endothelial cocultures. J Am Soc Nephrol 2002; 13: 2027−2036.
8.	KITAMURA S, MAESHIMA Y & SUGAYA T et al. Transforming growth factor 1 induces vascular endothelial growth factor expression in murine proximal tubule cells. Nephron Exp Nephrol 2003; 95: e79−e86.
9.	SHI Y & MASSAGUE J. Mechanisms of TGF- signalling from cell membrane to the nucleus. Cell 2003; 113: 685−700.
10.	NAKAGAWA T, LI JH & GARCIA G et al. TGF- induces proangiogenic and antiangiogenic factors via parallel but distinct Smad pathways. Kidney Int 2004; 66: 605−613.
11.	FONG G-H, ZHANG L, BRYCE D-M & PENG J. Increased haemangioblast commitment, not vascular disorganization, is the primary defect in flt-1 knock-out mice. Development 1999; 126: 3015−3025.