Portal de Periódicos da CAPES

Interstitial fibrosis: tubular hypothesis versus glomerular hypothesis

2008; Elsevier BV; Volume: 74; Issue: 10 Linguagem: Inglês

10.1038/ki.2008.421

1523-1755

Erik Christensen, P Verroust,

Renal Diseases and Glomerulopathies

The pathogenesis of renal interstitial fibrosis leading eventually to renal failure is highly debatable. Whereas the so-called tubular hypothesis, involving an increased tubular uptake of potentially toxic substances that induce a variety of cytokines, growth factors, and profibrogenic factors, is based to a large extent on cell-culture studies, the glomerular hypothesis is based mainly on careful morphological observations. Unraveling the pathways appears to be extremely complex, but in vivo studies appear to offer the most reliable results. The pathogenesis of renal interstitial fibrosis leading eventually to renal failure is highly debatable. Whereas the so-called tubular hypothesis, involving an increased tubular uptake of potentially toxic substances that induce a variety of cytokines, growth factors, and profibrogenic factors, is based to a large extent on cell-culture studies, the glomerular hypothesis is based mainly on careful morphological observations. Unraveling the pathways appears to be extremely complex, but in vivo studies appear to offer the most reliable results. Most if not all glomerular diseases involving extracapillary injury progressively develop extensive fibrotic processes, leading to nephron destruction and terminal renal failure. Two hypotheses have been put forward to account for this evolution. The first proposes that the primary event is tubular: the increased amount of protein that gains access to the proximal tubule, which results in increased protein trafficking in the proximal tubule cells, is toxic for the cells, thus triggering a number of inflammatory and fibrotic pathways. The second proposes that the primary event is glomerular: the formation of glomerular crescents leads to encroachment on the glomerular-tubular junction and subsequent tubular degeneration. Two recent studies provide additional data in this context. Motoyoshi et al.1.Motoyoshi Y. Matsusaka T. Saito A. et al.Megalin contributes to the early injury of proximal tubule cells during nonselective proteinuria.Kidney Int. 2008; 74: 1262-1269Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar (this issue) induced massive glomerular proteinuria in a mouse model, mating mosaic megalin kidney knockout mice with a transgenic mouse, NEP25, in which podocytes expressing the transgene CD25 can be damaged by exposure to the immunotoxin LMB2 binding to CD25.2.Matsusaka T. Xin J. Niwa S. et al.Genetic engineering of glomerular sclerosis in the mouse via control of onset and severity of podocyte-specific injury.J Am Soc Nephrol. 2005; 16: 1013-1023Crossref PubMed Scopus (213) Google Scholar Ten days after exposure to LMB2, these mice showed mild glomerular injury, and only 5% of the glomeruli showed modest to severe changes. Similarly, the mice showed mild tubulointerstitial changes with occasional tubule dilatation, protein casts, and rare fibrosis. The megalin-expressing cells preferentially accumulated albumin and immunoglobulin, and these cells also showed upregulation in the expression of the tubule injury markers heme oxygenase-1, monocyte chemoattractant protein-1 (MCP-1), and apoptosis. The inevitable conclusion of the authors is that the increased reabsorption of proteins, notably high-molecular weight proteins, triggers events that can lead to tubule injury. In a similar study, Theilig et al.3.Theilig F. Kriz W. Jerichow T. et al.Abrogation of protein uptake through megalin-deficient proximal tubules does not safeguard against tubulointerstitial injury.J Am Soc Nephrol. 2007; 18: 1824-1834Crossref PubMed Scopus (84) Google Scholar reached an apparently opposite conclusion. In this study, glomerulonephritis was induced in mosaic kidney-specific megalin knockout mice by injection of anti-mouse glomerular basement membrane serum, and the mice were analyzed 18 days later. The mice developed crescentic glomerulonephritis and tubulointerstitial disease with tubular dilation and casts, but also with atrophy, degeneration, and collapse. From a morphological analysis, the authors concluded that the tubulointerstitial changes were related directly to glomerular changes extending into the proximal tubular junction. They also concluded that excessive uptake of protein in megalin-expressing segments had no major impact on the development of tubulointerstitial disease. The authors suggested that the findings that cell proliferation and the inflammatory proteins transforming growth factor-β (TGF-β), intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and endothelin-1 were increased in megalin-expressing cells might be due to stress-protective functions rather than disease-induced mechanisms. Before we discuss the apparent discrepancy between the two studies, it is worth taking a brief look at the arguments underlying each hypothesis. The tubular hypothesis can be traced to a paper by Risdon and co-workers,4.Risdon R.A. Sloper J.C. De Wardener H.E. Relationship between renal function and histological changes found in renal-biopsy specimens from patients with persistent glomerular nephritis.Lancet. 1968; 2: 363-366Abstract PubMed Google Scholar). who first showed in human renal biopsies 40 years ago that progression to renal failure correlated more closely with tubular than with glomerular damage. Further evidence was derived from the close correlation between the levels of proteinuria and disease progression (reviewed by Abbate et al.5.Abbate M. Zoja C. Remuzzi G. How does proteinuria cause progressive renal damage?.J Am Soc Nephrol. 2006; 17: 2974-2984Crossref PubMed Scopus (590) Google Scholar). The bulk of evidence in favor of a toxic role of reabsorbed proteins was obtained in vitro with the use of proximal tubular cells in culture. When exposed to a variety of plasma proteins, proximal tubular cells secrete cytokines such as endothelin-1, MCP-1, RANTES, interleukin-8, and fractalkine. Secreted cytokines appear to be released essentially at the basolateral pole, suggesting that they may indeed be involved in interstitial inflammation. They probably recruit and stimulate interstitial macrophages, which, under the influence of growth factors and profibrogenic factors such as TGF-β, transform interstitial cells into myofibroblasts. TGF-β may also be secreted by proximal tubular cells. The mechanisms of protein uptake have not been studied in detail, and, in particular, the expression of megalin and cubilin in cell-culture studies is often not documented. In contrast, the mechanisms involved in cytokine induction have been actively explored. Exposure of various cells to albumin activates nuclear factor-κB (NF-κB) via a pathway involving protein kinase Cdependent generation of peroxide, mitogen-activated protein kinases (including p38, ERK1, and ERK2), and STAT. As Figure 15.Abbate M. Zoja C. Remuzzi G. How does proteinuria cause progressive renal damage?.J Am Soc Nephrol. 2006; 17: 2974-2984Crossref PubMed Scopus (590) Google Scholar illustrates, autocrine loops may exist, amplifying the inflammatory response. Finally, glomerular-produced growth factors and/or cytokines may reach the tubule. In vitro, these cytokines upregulate their receptors and induce basal secretion of MCP-1 and RANTES. In vivo evidence relating increased protein trafficking to interstitial inflammation and fibrosis is limited. Production of MCP-1 and activation of the NF-κB pathway have been described in five-sixths nephrectomy and passive Heymann's nephritis. Similar observations have been reported in patients' biopsies. It has also been reported that knockdown of key components of the putative pathway (for example, MCP-1 or NF-κB) improved interstitial inflammation. These experiments establish a link between interstitial inflammation and cytokine production, but they do not provide evidence directly linking these findings to proteinuria. The latter relies on pharmacological studies using angiotensin-converting enzyme inhibitors, which decrease proteinuria, inhibit the production of MCP-1 and the activation of NF-κB, and limit the accumulation of cells in the interstitium. The glomerular hypothesis (reviewed by Kriz and LeHir6.Kriz W. LeHir M. Pathways to nephron loss starting from glomerular diseases: insights from animal models.Kidney Int. 2005; 67: 404-419Abstract Full Text Full Text PDF PubMed Scopus (378) Google Scholar), sometimes referred to as the overload hypothesis, relies almost entirely on careful morphological analysis of various models of glomerular diseases. It proposes that inflammatory or degenerative processes of the glomerulus invariably decrease podocyte density, resulting in naked areas of glomerular basement membrane to which parietal epithelial cells may adhere. Such adhesions of the tuft to the glomerular capsule result in misdirected filtration through a podocyte-deprived filtration barrier. This results in the formation of the so-called proteinaceous crescents, which progressively transform into cellular and fibrocellular crescents and encroach on the glomerular tubular junction, resulting in atubular glomeruli and aglomerular tubules (Figure 1).6.Kriz W. LeHir M. Pathways to nephron loss starting from glomerular diseases: insights from animal models.Kidney Int. 2005; 67: 404-419Abstract Full Text Full Text PDF PubMed Scopus (378) Google Scholar Also, misdirected ultrafiltration may indeed occur and dissect tubular epithelial cells away from the tubular basement membrane. Key for this theory is the observation that a preserved tubule within damaged parenchymal and stromal tissue can be traced to a normal glomerulus. Dynamic experiments to further document this pathway are not easy to conceive. It has, however, been shown with the use of various tracers that misdirected ultrafiltration did occur.7.Kriz W. Hartmann I. Hosser H. et al.Tracer studies in the rat demonstrate misdirected filtration and peritubular filtrate spreading in nephrons with segmental glomerulosclerosis.J Am Soc Nephrol. 2001; 12: 496-506PubMed Google Scholar Of course, the two hypotheses are not mutually exclusive, and the objective at present is to define their role in a given pathological context. The paper by Motoyoshi et al.1.Motoyoshi Y. Matsusaka T. Saito A. et al.Megalin contributes to the early injury of proximal tubule cells during nonselective proteinuria.Kidney Int. 2008; 74: 1262-1269Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar and the paper by Theilig et al.3.Theilig F. Kriz W. Jerichow T. et al.Abrogation of protein uptake through megalin-deficient proximal tubules does not safeguard against tubulointerstitial injury.J Am Soc Nephrol. 2007; 18: 1824-1834Crossref PubMed Scopus (84) Google Scholar present data of significance in this context. First, both clearly show that megalin-dependent endocytosis significantly impacts the production of various markers of tubular injury, including proinflammatory/profibrotic components or adhesion molecules such as MCP-1 and heme oxygenase for Motoyoshi et al.1.Motoyoshi Y. Matsusaka T. Saito A. et al.Megalin contributes to the early injury of proximal tubule cells during nonselective proteinuria.Kidney Int. 2008; 74: 1262-1269Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar and TGF-β, ICAM-1, VCAM-1, and endothelin-1 for Theilig et al.3.Theilig F. Kriz W. Jerichow T. et al.Abrogation of protein uptake through megalin-deficient proximal tubules does not safeguard against tubulointerstitial injury.J Am Soc Nephrol. 2007; 18: 1824-1834Crossref PubMed Scopus (84) Google Scholar This direct demonstration is important because previous studies relied on the antiproteinuric effects of angiotensin-converting enzyme inhibitors, which, as was previously pointed out, may also prevent the non-hemodynamic effects of angiotensin II on proximal tubule cells, such as proliferation and TGF-β production. Interestingly, the effects observed, notably on MCP-1, are not identical in the two studies. The reason for this is unknown. However, the experimental design of both papers does not give the authors the opportunity to relate the effects observed to the nature of the proteins internalized, which obviously varies depending on the type of glomerular lesion. Thus, albumin alone may be active, but components bound by albumin (free fatty acids, for instance) may also be important. It has also been proposed that transferrin had greater effects, and, perhaps most important, that complement was essential.8.Hsu S.I. Couser W.G. Chronic progression of tubulointerstitial damage in proteinuric renal disease is mediated by complement activation: a therapeutic role for complement inhibitors?.J Am Soc Nephrol. 2003; 14: S186-S191Crossref PubMed Google Scholar,9.Abbate M. Zoja C. Corna D. et al.Complement-mediated dysfunction of glomerular filtration barrier accelerates progressive renal injury.J Am Soc Nephrol. 2008; 19: 1158-1167Crossref PubMed Scopus (53) Google Scholar In addition, the cytokines, growth factors, and growth factor precursors that gain access to the proximal tubule will also vary with the type of glomerular lesion. One important point in the study by Theilig et al.3.Theilig F. Kriz W. Jerichow T. et al.Abrogation of protein uptake through megalin-deficient proximal tubules does not safeguard against tubulointerstitial injury.J Am Soc Nephrol. 2007; 18: 1824-1834Crossref PubMed Scopus (84) Google Scholar is the attempt to analyze both hypotheses in the same model. However the anti-glomerular basement membrane model selected is a rapidly progressive glomerulonephritis characterized by the early formation of florid crescents. Such a model will obviously emphasize the glomerular hypothesis. In addition, the short course of the disease and the only partial inactivation of megalin do not allow a full analysis of the tubular component. The study of Motoyoshi et al.,1.Motoyoshi Y. Matsusaka T. Saito A. et al.Megalin contributes to the early injury of proximal tubule cells during nonselective proteinuria.Kidney Int. 2008; 74: 1262-1269Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar on the other hand, uses a model that should be more suitable, as the induction and the severity of the glomerular disease can be controlled. The study is limited by the short time span, the partial inactivation of megalin, and the minimal study of the glomerular lesions. The question of whether tubulointerstitial changes induced by glomerular disease are a direct consequence of the glomerular disease spreading into the initial proximal tubule, causing obstruction of the proximal tubule and subsequent degeneration, or whether they are due to enhanced protein reabsorption and toxic effects on proximal tubule function inducing enhanced production of inflammatory proteins, has not been answered fully by these two studies. It is not possible, either, to conclude that the consequences of megalin-dependent protein endocytosis have a protective role. It is, however, interesting to note that Motoyoshi et al.1.Motoyoshi Y. Matsusaka T. Saito A. et al.Megalin contributes to the early injury of proximal tubule cells during nonselective proteinuria.Kidney Int. 2008; 74: 1262-1269Abstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar found increased heme oxygenase-1 production, which is likely to protect tubules from the fibrogenic effects of TGF-β. As an example, the complexity of the system is illustrated by the fact that heme oxygenase-1 activates NF-κB, which is generally considered pathogenic but may, in some instances, be protective.10.Kie J.H. Kapturczak M.H. Traylor A. et al.Heme oxygenase-1 deficiency promotes epithelial-mesenchymal transition and renal fibrosis.J Am Soc Nephrol. 2008; 19: 1681-1691Crossref PubMed Scopus (81) Google Scholar In our opinion, these papers can serve to design future experiments to specifically clarify the role of protein trafficking in the progression to renal failure. Carefully selected models of graded glomerular injury are available, and models of megalin and/or cubilin inactivation are being developed. Because of the complex nature of the events, dissection of the influence of individual cytokines, growth factors, and so on will probably be best carried out with the use of one-by-one knock-in/knockout animal models. The authors declared no competing interests. This work was supported in part by the University of Aarhus, and the European Community (EuReGene LSHG-CT-2004-005085).