Pathways to nephron loss starting from glomerular diseases—Insights from animal models

Kidney International, Vol. 67 (2005), pp. 404–419

PERSPECTIVES IN BASIC SCIENCE

Pathways to nephron loss starting from glomerular diseases—Insights from animal models WILHELM KRIZ and MICHEL LEHIR Institute of Anatomy and Cell Biology, University of Heidelberg, Heidelberg, Germany; and Institute of Anatomy, University of Zurich, Zurich, Switzerland

far, no convincing evidence has been published that in primary glomerular diseases fibrosis is harmful to healthy nephrons. The potential of glomerular injuries to regenerate or to be repaired by scaring is limited. The only option for extracapillary injuries with tuft adhesion is repair by formation of a segmental adherent scar (i.e., segmental glomerulosclerosis).

Pathways to nephron loss starting from glomerular diseases— Insights from animal models. Studies of glomerular diseases in animal models show that progression toward nephron loss starts with extracapillary lesions, whereby podocytes play the central role. If injuries remain bound within the endocapillary compartment, they will undergo recovery or be repaired by scaring. Degenerative, inflammatory and dysregulative mechanisms leading to nephron loss are distinguished. In addition to several other unique features, the dysregulative mechanisms leading to collapsing glomerulopathy are particular in that glomeruli and tubules are affected in parallel. In contrast, in degenerative and inflammatory diseases, tubular injury is secondary to glomerular lesions. In both of the latter groups of diseases, the progression starts in the glomerulus with the loss of the separation between the tuft and Bowman’s capsule by forming cell bridges (parietal cells and/or podocytes) between the glomerular and the parietal basement membranes. Cell bridges develop into tuft adhesions to Bowman’s capsule, which initiate the formation of crescents, either by misdirected filtration (proteinaceous crescents) or by epithelial cell proliferation (cellular crescents). Crescents may spread over the entire circumference of the glomerulus and, via the glomerulotubular junction, may extend onto the tubule. Two mechanisms concerning the transfer of a glomerular injury onto the tubulointerstitium are discussed: (1) direct encroachment of extracapillary lesions and (2) protein leakage into tubular urine, resulting in injury to the tubule and the interstitium. There is evidence that direct encroachment is the crucial mechanism. Progression of chronic renal disease is underlain by a vicious cycle which passes on the damage from lost and/or damaged nephrons to so far healthy nephrons. Presently, two mechanisms are discussed: (1) the loss of nephrons leads to compensatory mechanisms in the remaining nephrons (glomerular hypertension, hyperfiltration, hypertrophy) which increase their vulnerability to any further challenge (overload hypothesis); and (2) a proteinuric glomerular disease leads, by some way or another, to tubulointerstitial inflammation and fibrosis, accounting for the further deterioration of renal function (fibrosis hypothesis). So

The decline in renal function in chronic renal failure is caused by the progressive loss of viable nephrons. Surprisingly few studies have dealt with the question of how a nephron undergoes irreversible degeneration during a specific kidney disease. It is widely believed that excessive matrix production, either in the glomerular mesangium or the tubulointerstitium, plays a crucial role. However, the evidence for such a mechanism is entirely correlative in nature; a causal relationship, in our view, has never been established. In humans and in laboratory animals, the majority of diseases which progress to chronic renal failure start at the glomerulus. This review will deal with pathways of progression of a primary glomerular disease to complete nephron degeneration. Pathway is defined here as a series of consecutive and causally related events starting with a healthy nephron and ending with its loss. We will mainly present insights from animal models. Though there are fundamental consistencies and ample similarity in details with glomerular diseases in humans, we will, in general, not expressly mention the latter; this would break up the frame of this review. We have excluded, first, the diseases beginning at preglomerular vessels (i.e., all kinds of vaso-occlusive diseases leading to ischemic events downstream) and, second, the diseases which begin within the interstitium (primary interstitial diseases) or with direct damage to the tubule (by toxic, ischemic and/or obstructive mechanisms).

Key words: glomerular diseases, transfer onto the tubulointerstitium, progression to chronic renal failure, repair.

GLOMERULAR DISEASES Glomerular diseases comprise a large variety of individual entities. Our working hypothesis is that all these primary diseases converge into few pathways of disease progression.

Received for publication May 6, 2004 and in revised form July 15, 2004 Accepted for publication September 8, 2004
C

2005 by the International Society of Nephrology 404

Kriz and LeHir: Pathways to nephron loss

When talking about glomerular pathomechanisms, it is most useful to distinguish between the endocapillary compartment, which consists of the capillaries and the mesangium and, separated by the glomerular basement membrane (GBM), the extracapillary compartment, which consists of podocytes, Bowman’s space and Bowman’s capsule. The latter comprises the parietal epithelium and the parietal basement membrane (PBM). Basically, the endocapillary compartment is a capillary network with surrounding mesangial cells; thus it comes without surprise that pathological processes share many features with those in other capillary beds (e.g., inflammatory processes, wound healing). In contrast, the extracapillary compartment is unique and absolutely specific to the glomerulus; this holds true for the pathologic developments in this compartment. As long as a glomerular disease remains restricted to the endocapillary compartment, it is subject to restitution or repair even if the lesions are massive (see below). According to what is seen in animal models, progression of a glomerular disease starts with the involvement of the extracapillary compartment. Every extracapillary process interferes with the filtration barrier and the separation between the tuft and Bowman’s capsule, leading, in general locally by forming a tuft adhesion to Bowman’s capsule, to the obliteration of Bowman’s space. Summarized in a synopsis (Fig. 1), three different mechanisms may be distinguished as being responsible for the progression of a glomerular disease toward nephron loss. These are degenerative, inflammatory, and, dysregulative mechanisms; essential structural features are depicted in Figure 2. These three mechanisms represent prototypes of damage development in the glomerulus. In an actual case (as will be discussed later in more detail) they may well be mixed. For instance, a “cellular lesion” (known in human pathology as a dense assembly of podocytes on the surface of an injured tuft portion; [27]), in our view, represents a local manifestation of the dysregulative pathway that may well be encountered in a predominantly degenerative glomerular disease. Dysregulative mechanisms The dysregulative mechanisms are clearly different from both others, the degenerative and inflammatory ones. The unique histologic pattern, now named collapsing glomerulopathy, is most frequently encountered in patients with human immunodeficiency virus (HIV) nephropathy, but idiopathic cases occur as well [28, 29]. It is debated whether the “cellular lesion” in focal segmental glomerulosclerosis (FSGS) belongs to the spectrum of collapsing glomerulopathy or represents yet another variant of FSGS [28–31]. Transgenic models (see Fig. 1) exhibiting a strikingly similar histopathology have led to

405

Fig. 1. Progression of extracapillary injuries.

insights into mechanisms leading to this lesion. The dysregulative pathway may be summarized as follows [1–3, 8, 32–34]. HIV infection or some other so far unknown intracellular event cause regulatory defects in glomerular and tubular epithelial cells. In the glomerulus, dysregulation of podocytes leads to the loss of the specific cell shape and to uncontrolled proliferation [33, 35]. The now cuboidal cells display many features of immature podocytes during nephrogenesis. These dedifferentiated cells fill Bowman’s space and form “pseudocrescents.” Furthermore, a defect in the regulation of vascular endothelial growth factor (VEGF) synthesis by podocytes

406

Kriz and LeHir: Pathways to nephron loss

Fig. 2. Esssential structural features of the dysregulative (A), the degenerative (B), and the inflammatory (C) pathway to nephron degeneration. The glomerular basement membrane (GBM) is shown in black, mesangial cells in a bird-feet pattern, podocytes are densely stippled, and macrophages are loosely stippled. Parietal cells are shown in dark brown, proximal tubule cells in pale brown, and fibroblasts in green. Abnormal filtrate spreading is indicated in yellow. See text for further explanation.

[7, 36] results in endothelial damage followed by the collapse of glomerular capillaries, and ultimately of an entire lobule or even the entire tuft. In tubules the regulative defects lead to false sorting of membrane proteins and progressive formation of microcysts [2, 3], resulting in tubular degeneration and tubulointerstitial injury with final destruction of the entire cortex. Congenital diseases in humans exhibiting this kind of nephron degeneration (e.g., Denys-Drash syndrome) have long been known; they are based on mutations in the Wilms’ tumor gene [37, 38]. Degenerative mechanisms The degenerative and the inflammatory mechanisms share several important histopathologic features; in particular both lead to crescent formation as the crucial step to nephron loss. In the degenerative models (see Fig. 1), apart from the toxic models, podocytes are exposed to persistent long-term challenges, including glomerular hypertension, metabolic disturbances as well as abnormal receptor stimulation either by growth factors or vasoactive compounds. Functionally, these challenges account for the loss of size selectivity of the glomerular barrier, leading to excessive protein leakage into tubular urine. Structurally, in addition to hypertrophy, a stereo-

typed pattern of lesions are encountered (foot process effacement as the most common lesion, cell-body attenuation, pseudocyst formation, accumulation of absorption droplets, sometimes “microvillous transformation”) followed finally by detachment [39–41]. These events cause a continuous loss of podocytes and, frequently together with an hypertrophic growth of the tuft, lead to a decrease in podocyte density which is seen under various circumstances [13, 16, 20, 42–45]. Since podocytes in the adult are incapable of regenerative cell replication, lost podocytes cannot be replaced by new cells [19, 42, 46]. The hypertrophy of the remaining podocytes is the only way to maintain the integrity of the filtration barrier. The additional workload on these cells increases their vulnerability; a vicious circle is initiated. The podocytes eventually are too few to maintain a complete cover of the tuft. Naked areas of GBM occur, to which parietal epithelial cells can adhere. Such a parietal cell bridge goes necessarily along with the formation of a local gap within the parietal epithelium, including the formation of naked areas of PBM. By establishing further cell bridges between the PBM and the GBM adjacent to the first and deposition of some matrix between the bridging cells, finally an adhesion of the tuft to Bowman’s capsule is established [39, 46].

Kriz and LeHir: Pathways to nephron loss

407

rich filtrate into the interstitium instead of Bowman’s space [47, 48]. In response, interstitial fibroblasts create a multilayered cellular patch around the focus of misdirected filtration, limiting the entry of fluid into the interstitium. This results in the formation of a crescent-shaped space filled with a proteinaceous fluid overarching the adhesion (proteinaceous crescent) (Fig. 3); hyaline material also frequently accumulates within the adherent portion of the tuft. Actually, the hyalinosis [54] inside the GBM and the proteinaceous crescent outside are the two sides of the same coin (i.e., of misdirected filtration through a podocyte-deprived filtration barrier: Plasma proteins accumulate on both sides. As outlined in Figure 3, a nascent crescent consists of three portions. Spreading of the proteinaceous fluid within the middle compartment into all directions leads to an expansion of the proteinaceous crescent on the outer surface of the glomerulus and eventually also of the tubule (see below). In the case that this process comes to a standstill, frequently at the border of a lobule, a repair process by scaring may lead to a synechia (i.e., to segmental glomerulosclerosis (Fig. 4).

Fig. 3. Crescents.

The adherent tuft portion generally contains at least one capillary loop that bulges with its naked (podocytedeprived) GBM surface toward the core of the adhesion. As a consequence, such capillaries deliver their protein-

Inflammatory mechanisms There are numerous types of glomerulonephritis in humans and also numerous animal models. We will not discuss specific diseases. Instead, in keeping with the subject of this review, we will focus on the most frequent pathway to nephron loss in glomerulonephritis (i.e., formation of a cellular crescent). In our presentation, we rely largely on our own work, which, in addition to unpublished work, includes models of anti-GBM nephritis in rats and mice, “classic” Masugi nephritis, pauci-immune-glomerulonephritis (SCG/kj mouse) and Thy-1–mediated mesangioproliferative glomerulonephritis (see Fig. 1). We have carefully considered studies from other investigators dealing with histopathology of glomerulonephritis in animal models [52, 60–65] as well as in humans [66–73]. A glomerular inflammation starts in the endocapillary compartment, affecting the capillaries and the mesangium. It may take a mild course, remaining bound within the endocapillary compartment. This is associated with an early turn to restitution (see below). The endocapillary process may take an aggressive course, leading to breaks in the GBM and exudation into Bowman’s space. Breaks in the GBM and spreading of the inflammation in Bowman’s space constitute a dramatic injury [49]. The glomerulus will most likely die and the nephron be lost. Formation of cellular crescents is probably associated with a moderate level of severity of the inflammation ranging in between the mild and the aggressive course described above. The way that a crescent develops from an endocapillary process has long been an enigma. It is widely believed that crescents develop within Bowman’s

408

Kriz and LeHir: Pathways to nephron loss

Fig. 4. Sclerosis, glomerulosclerosis, collapsing glomerulopathy.

space as a result of breaks in the GBM and egress of blood borne compounds (among them fibrinogen) into Bowman’s space [52, 62, 63]. However, in several experimental studies, exudates were not conspicuous at early stages of crescent formation [22, 25, 60, 61, 64]. Our observations in experimental models of glomerulonephritis suggest that the crucial role in crescent formation is played by epithelial cells that establish cell bridges between the tuft and Bowman’s capsule. This is in agreement with many studies in humans and in experimental animals, showing that early crescents include cells of epithelial origin (see below). An endocapillary inflammation induces a stereotyped pattern of morphologic alterations in podocytes. Most of these changes are frequent in noninflammatory diseases as well: simplification of foot processes, focal detachment from the basement membrane, increased incidence of intracellular vacuoles, and microvilli-like microprotrusions. The most specific feature of inflammatory diseases are coarse and irregular microprojections on the apical aspect

of the cell body. In the inflammatory models which we have examined (see Fig. 1), this phenotype of podocytes was associated with the formation of podocyte bridges between the tuft and Bowman’s capsule. Podocyte bridges are by definition made up of cells which adhere to both the GBM and the PBM. It is not known by which mechanism a cell process of a podocyte gets across the parietal cell layer. It is likely that podocyte processes disrupt the junctional complexes between two parietal cells and fix to the PBM. At the PBM, they frequently enlarge to lamellipoda-like structures, hereby dislodging the parietal epithelial cells [22, 25]. Frequently, cells adjacent to a podocyte bridge develop further bridges, including those of parietal epithelial cells that fix to the GBM [25]. After depositing basement membrane-like matrix between the cells, the entire bridging structure interconnecting the GBM and the PBM represents a tuft adhesion to Bowman’s capsule. The altered location of both types of glomerular epithelial cells together with loss of usual junctional

Kriz and LeHir: Pathways to nephron loss

409

Fig. 5. Transfer of a glomerular injury onto the tubulointerstitium.

connections [22] obviously profoundly alters their behavior. They proliferate and spread, resulting in the formation of a cellular crescent. It is noteworthy that the spreading of crescent cells takes place inside the PBM, but outside a more or less well preserved parietal epithelium, thus (like in a proteinaceous crescent, discussed above) in a newly emerging space between the parietal epithelium and its basement membrane; this portion represents the crescent proper (Fig. 3). Apart from the connecting stalk between the tuft and the crescent, a narrow urinary space subsists between the crescent and the tuft. Of note, this feature is visible only in perfusion-fixed kidneys. Since the starting point for crescent formation is a podocyte bridge between a capillary loop and the PBM, it is not surprising to find at least one capillary loop included in the adherent tuft portion of every crescent. In

an advanced stage, these loops are generally hyalinized, but perfused capillaries may also be found that obviously deliver their filtrate into the crescent. Thus, by the same mechanism as described above in degenerative models, proteinaceaous crescents may be formed in inflammatory models as well. Mixed crescents (cellular/proteinaceous) are frequent. Young crescents contain mainly epithelial cells as seen in animal models [22, 60, 61, 64], as well as in humans [66, 68, 70–74]. As shown recently, a significant fraction of epithelial cells in crescents are derived from podocytes [64, 75], which, during their displacement and proliferation, have lost all specific podocyte markers. This phenotype represents, in addition to collapsing glomerulopathy, another example of the proliferative ability of dedifferentiated podocytes. Podocytes contribute only to

410

Kriz and LeHir: Pathways to nephron loss

a fraction of the epithelial cells of a crescent [75]. The rest probably emerges from proliferation of parietal epithelial cells as already suggested in older work on human glomerulonephritis [70, 73] and on a murine model [60]. In agreement with the epithelial origin of the cells, the intercellular matrix of early (“cellular”) crescents consists of an amorphous basement membrane-like material in both human [68, 76] and murine glomerulonephritis [22] in addition to varying proportions of proteinaceous fluid. Macrophages are a further component of crescents in humans [66, 68, 69, 71, 73] and in animal models [63]. According to our observations, macrophages first accumulate within the cap of proliferating interstitial tissue that covers a nascent crescent from the outside (Fig. 3). In early crescents, this interstitial reaction is separated from the crescent proper by the intact PBM. Later, macrophages increasingly intermingle with epithelial cells within the crescent proper, most likely by invasion through breaks in the PBM as shown in humans [70] and in a model of glomerulonephritis in the rat [52]. Like with proteinaceous crescents, cellular crescents may progress to more severe damage, eventually to nephron loss (next chapter) or they may undergo repair by scarring, terminating in segmental glomerulosclerosis (Fig. 4). TRANSFER OF A GLOMERULAR INJURY ONTO THE TUBULOINTERSTITIUM Glomerular damage is generally accompanied by tubulointerstitial damage and there is widespread agreement that the former causes the latter. However, there is disagreement concerning the underlying mechanisms. As summarized in the synopsis shown in Figure 5, several mechanisms are under discussion. The most popular notion is that protein leakage in injured glomeruli into tubular urine leads to excessive protein reabsorption in proximal tubules, resulting in tubulointerstitial injury. Recently, mechanisms have been uncovered that lead to a direct encroachment of extracapillary lesions onto the tubule via the glomerulotubular junction; they will be addressed first. Encroachment of a glomerular injury onto the glomerulotubular junction Two mechanisms have been uncovered. First, there is an extension of a proteinaceous crescent onto the outer aspect of the proximal tubule. As a consequence of persistent misdirected filtration, the proteinaceous filtrate, via the glomerulotubular junction, expands in the space between the tubular epithelium and the tubular basement membrane and may spread within this space along the entire proximal convolution. This process is predominantly associated with degenerative glomerular diseases [47, 48],

Table 1. Crescentic glomerulonephritis in mice (N = 265 glomeruli examined in section series) Number 168 glomeruli without crescents 97 crescentic glomeruli 57 crescents do not extend in tubular neck 40 crescents obstruct the tubular neck

Description Corresponding tubules show normal structure Corresponding tubules show normal structure Corresponding tubules show features of degeneration

Serial section analysis of glomerulotubular connections on day 10 in the disease [23].

but may also be found as the dominant process in individual nephrons in inflammatory diseases [25]. The second mechanism is the encroachment of a growing cellular crescent upon the glomerulotubular junction. Thereby the initial segment of the proximal tubule gets incorporated into the crescent. This process is characteristically found in inflammatory models [23, 25, 26]. However, cellular proliferation is complemented frequently in inflammatory models by misdirected filtrate spreading, leading to mixed crescents. These two processes appear to be very frequent in experimental models under various circumstances and have also been shown to occur in human cases [48, 77, 101]. In a murine model of crescentic glomerulonephritis [23], the relevance for final nephron degeneration has been quantified by tracing 265 nephrons in serial sections (Table 1). Exclusively glomeruli with crescents and among them only those that extended onto the glomerulotubular junction were associated with tubular degeneration in various stages (see below). In our view, this represents very strong evidence for the concept that the tubular injury develops by encroachment of a glomerular injury onto the glomerulotubular junction. Both mechanisms, abnormal filtrate spreading and cellular overgrowth, may lead to the loss of the nephron. A nephron is functionally lost when urine flow is irreversibly impeded by tubular degeneration. The latter may occur directly or it may be preceded by tubular atrophy (both processes are described below in detail). Peritubular filtrate spreading, which is less aggressive compared to cellular overgrowth, may first lead to tubular atrophy. At this stage, the nephron is still in function and produces urine, even though tubular functions are necessarily impaired. Such nephrons may reach a status of some kind of repair, exhibiting a segmental adherent scar at the glomerulus and an atrophic portion of the proximal tubule. It is unknown how long these nephrons may survive. As seen in experimental models, the atrophic portion of the tubule, sooner or later, will proceed to degeneration, get obstructed and thus irreversibly damaged (see below). Cellular overgrowth of the glomerulotubular junction directly leads to tubular degeneration with early obstruc-

Kriz and LeHir: Pathways to nephron loss

tion generally affecting the glomerulotubular junction. Thus the nephron is functionally lost. As intermediate stages to complete structural loss, atubular glomeruli and aglomerular tubular remnants are generally encountered. The final decomposition and the removal of the tubular remnants always occur within a short time in a steadily progressing manner in conjunction with a local interstitial inflammation, whereas the final decomposition of the glomerular remnants may be delayed (see below). Atubular glomeruli thus represent a common finding in experimental as well as human diseases [10, 14, 15, 23, 25, 48, 101–107]. Tubular atrophy begins with the loss of the brush border and the basal interdigitating cell processes, including their mitochondria. In the end, the complex proximal epithelium is transformed into a simple flat epithelium resembling the parietal epithelium of Bowman’s capsule [25, 48, 108]. Functionally, this dramatically simplified epithelium does not appear to be capable of performing any of the specific transports of the proximal tubular epithelium. Tubular degeneration starts with obstruction of the tubule, generally at or close to the glomerulotubular junction. Deprived of a luminal load, the tubules collapse, the cells lose their arrangement as a polarized epithelium and transform into solid cords of polygonal cells. Lysosomal elements accumulate; finally the cells die. The empty remains of the tubular basement membrane become wrinkled and infiltrated with inflammatory cells. Cellular and matrix residues are finally removed by macrophages and replaced by fibrous tissue [23, 25, 48]. Atubular glomeruli all have a tuft-to-capsule connection. They may either collapse or develop cystic expansions of Bowman’s capsule [25, 48]. Collapsed glomeruli appear to undergo a slow (compared with that of tubules) process of decomposition. The characteristics, podocyte dedifferentiation with loss of the cytoskeleton, detachment and disappearance of adjacent capillaries, collapse of the “capillary-deprived” GBM, are reminiscent of tuft degeneration in collapsing glomerulopathy [32]. Cystic expansion of Bowman’s space appears to develop in glomeruli with some ongoing filtration and in which Bowman’s capsule does not withstand expansion. These glomerular remnants appear to survive for long periods, possibly lifelong [25]. The decomposition of nephron remnants induces a local interstitial inflammation [23, 25, 48]. Inflammatory cells immigrate and fibroblasts transform into a-smooth muscle actin (a-SMA)-positive cells. After removal of the nephron remnants, this proliferative tissue transforms into fibrous tissue. From a teleologic point of view, these processes are responsible for the removal of the nephron remnants and for the filling of the space left by degenerated tubules. Since at the onset of this process the nephron is already irreversibly damaged, in our view, this

411

process guarantees the preservation of an intact tissue architecture. Protein leakage into tubular urine It has been proposed that continuous protein leakage through an impaired glomerular filter into tubular urine is not merely an indicator of severity of disease but also a trigger of progression. The “proteinuria hypothesis” has been the subject of many reviews [90–93, 109–113]. Several mechanisms have been proposed as to how filtered proteins (including protein-bound substances) may lead to tubulointerstitial damage. Since they have been amply presented in the above-mentioned reviews, we present a synopsis of the various proposals (Table 4). As seen there, they may be subdivided into (1) effects of excessive reabsorption and lysosomal degradation of proteins, (2) direct effects of toxic compounds, and (3) secondary effects of proinflammatory and profibrotic mediators produced by tubular cells; the resulting inflammation/fibrosis is supposed to be responsible for the loss of nephrons. The third mechanism has received the broadest support so far. In our models, we found no convincing evidence for harmful effects of filtered proteins on tubular cells [21, 23, 25, 47]. In contrast, even in cases of severe glomerular injury, the proximal tubule remained healthy in appearance as long as the glomerular injury did not extend onto the glomerulotubular junction (Table 1) (i.e., in cases with encroachment of extracapillary glomerular lesions). Since the question as to the damaging effects of protein leakage and interstitial inflammation presumably derived thereof not only relates to the particular nephron with protein leakage but also to neighboring so far healthy nephrons, thereby becoming an autonomous factor of progression, we will discuss this in the next section. PROGRESSION OF THE DISEASE The progressive decline in renal function in chronic renal disease is underlain by the continuous loss of nephrons, independent of whether the original disease persists. Thus, once a certain degree of injury is established, the disease enters into a vicious cycle. Indeed, in most cases of chronic renal disease, the histopathologic pattern is focal from the beginning. Severely damaged nephrons are found close to rather normally looking ones. It might thus be expected that a fraction of the nephrons would maintain a stable function when the primary disease heals or is therapeutically brought under control. However, clinical experience does not support that expectation and generally a progressive deterioration of renal function occurs. This raises the question as to a mechanism whereby nephron loss is deleterious to surviving nephrons. There are basically two hypotheses, overload and fibrosis, to explain how a damage may be passed on from one nephron to another.

412

Kriz and LeHir: Pathways to nephron loss

Fibrosis hypothesis According to this hypothesis the tubulointerstitial inflammation, which regularly develops in proteinuric diseases in conjunction with tubular degeneration, leads to progressive fibrosis accounting for the further deterioration of renal function [81, 90–93, 109, 113, 114]. This hypothesis was originally developed on the basis of three observations. First, several studies showed that the decline in renal function in chronic renal disease correlates more closely with interstitial fibrosis than with glomerular damage [92, 115–120]. Second, the level of proteinuria correlates well with the rate of progression in various diseases [91, 93, 109, 113, 121]. Third, exposure to high concentrations of proteins induces the production of proinflammatory and profibrotic factors by tubular cells [91, 93, 94, 111, 122]. Accordingly, it is assumed that protein overload in a tubule induces a local interstitial inflammation that eventually destroys not only that tubule but also neighboring tubules. This assumption is essential to every discussion about the impact of fibrosis on progression. In most publications on this subject, this question is not explicitly raised; it is simply assumed that fibrosis is harmful. A concrete hypothesis has never been formulated on how fibrosis might proceed to the loss of the nephrons. This brings us to the decisive questions on this issue: Does the focal tubulointerstitial inflammatory process that consistently occurs in glomerular diseases initiate nephron destruction? Is there a start of nephron destruction in the tubulointerstitium? The answer from all our studies of degenerative as well as inflammatory models of nephron degeneration is clearly “no.” An interstitial inflammation was only encountered in conjunction with the degeneration and removal of an already irreversibly damaged nephron [23, 25, 123]. By tracing in serial sections, it became obvious that tubular segments of healthy nephrons, even if fully surrounded by degenerating tubules of neighboring affected nephrons, preserved their usual features of a differentiated, normally functioning epithelium [12, 23, 25]. Thus, we found no evidence of progression of the disease at the level of the tubulointerstitium. On the contrary, by removing the remnants of nephrons and by replacing them with scars, the interstitial reaction assures that the degenerative mechanisms remain confined to the affected nephron. This conclusion agrees with experimental studies aiming at directly testing the effect of proteinuria and of proteinuria-induced interstitial inflammation on the progression of renal failure. Daily injections of bovine serum albumin for at least 1 week provokes a prominent diffuse infiltration of the peritubular interstitium with macrophages and lymphocytes [91, 124–126]. How-

ever, neither nephron loss nor a progressive reduction of renal function have been demonstrated in those studies. Adriamycin nephropathy induced in analbuminemic rats compared to wild-type rats [127, 128] did not show any difference between the two strains. In spite of a massive decrease in total proteinuria and of the lack of urinary albumin in mutant rats, the morphology of the tubulointestitium and the decrease in inulin clearance were not different between the two groups. This shows that tubular albumin is not a critical factor in producing the tubulointerstitial alterations in that model but, surely, does not rule out a specific pathogenetic role of some other proteins. The preservation of renal function in minimal change nephropathy in human as well as in several experimental studies mimicking this disease [14, 15, 126] speak against a major impact of proteinuria per se on the tubulointerstitial disease. In one of these models, namely in adriamycin nephropathy [14], the stage of pure minimal change nephropathy was followed, without significant increase in the degree of proteinuria, by the development of FSGS. Only at that stage, nephron loss, focal interstitial inflammation around the degenerating tubules and deterioration of renal function occurred. Serial sections revealed an association between the tubular degeneration and the encroachment of the extracapillary lesion onto the glomerulotubular junction. The evidence supporting a crucial contribution of protein leakage and of the subsequent interstitial reaction to progression is correlative in nature. It is grounded on two well-documented sets of clinical data. First, a correlation exists between the amount of excreted proteins and the pace of disease progression [109, 113, 121, 129]. However, the degree of glomerular damage is a confounding factor in that correlation: The more severe the glomerular injury, the higher the level of proteinuria but also the risk that glomerular damage directly encroaches upon the tubule and leads to nephron degeneration. Second, the decline in renal function in chronic renal failure correlates more closely with interstitial fibrosis than with glomerular damage (see above). This fact is taken as argument in favor of a genuine interstitial mechanism of progression. Thereby one overlooks a crucial fact: As animal models (see above) and studies in the aging human kidney show [130, 131], the remnants of a degenerative nephron, including glomeruli, may be completely removed and replaced by fibrous tissue. Thus, with the degeneration of nephrons, the interstitial damage score will increase, while the glomerular damage score will remain constant or even decrease due to complete disappearance of sclerotic glomeruli. Recent studies clearly show that the decline in renal function in chronic renal disease correlates best with the number of remaining nephrons [132]. Taken together, the correlations between proteinuria and interstitial fibrosis on the one hand and progression of

Kriz and LeHir: Pathways to nephron loss

the disease on the other do not allow any conclusion as to the pathways along which the nephron underwent destruction. Thus, the evidence for the fibrosis hypothesis is far from being conclusive. Overload hypothesis The core of this hypothesis reads as follows. Loss of nephrons leads to a work overload on the remaining nephrons followed by compensatory mechanisms (i.e., by hypertrophy), in particular by glomeruli exposing them to hypertension resulting in hyperfiltration and protein leakage. This leads sooner or later to glomerular damage that, according to the degenerative pathway to sclerosis discussed above, may proceed to nephron loss, thus closing a vicious cycle based on nephron loss [133–138]. The validity of this hypothesis has been repeatedly confirmed in the model of nephron mass reduction by nephrectomy or infarction. In several of those studies, systemic blood pressure as well as intraglomerular blood pressure were manipulated by various means. Depending on the protocol, the remnant nephrons compensated partially or fully for the decrease of nephron number by increasing single nephron GFR. Thereby, the intraglomerular blood pressure rose and glomerular hypertrophy was initiated. On the long run, a deterioration of renal function and development of glomerulosclerosis were observed. Intraglomerular blood pressure appears to be the main pathogenic trigger in this model [55, 133, 136, 139, 140], but a contribution of glomerular hypertrophy is likely [42, 135, 138, 141–143]. Thus, the evidence supporting the overload hypothesis is strong. However, many details are insufficiently understood, in particular the regulatory routes along which the compensatory hyperfunctions are elicited as well as how glomerular hypertension translates into protein leakage and glomerular damage [144]. In animal models, the hypertensive injury is specific; it consists of the vascular pole associated glomerulosclerosis that develops from a pressure dependent dilation of the first branches of the glomerular afferent arteriole [10, 145]. This suggests that the initiating failure consists of the incapability of the vessel wall to develop sufficient counterforces. Numerous studies [9, 146–148] support the notion that indeed the failure of podocytes to generate sufficient counterforces in response to high capillary pressure accounts for the start of sclerosis development. Under such conditions, podocytes up-regulate several genes, including osteopontin, which is known to be elevated in glomeruli in vivo in the deoxycorticosteroe acetate (DOCA)-salt model associated with glomerular hypertension [149]. Podocytes have stretch-sensitive ion channels and may respond along Ca++ dependent and independent signaling routes [150]. Podocytes have an-

413

giotensin II type 1(AT1) receptors [151] that become upregulated in response to stress [152]. Increased signaling via AT1 receptors is damaging to the podocyte, resulting in protein leakage and finally in sclerosis [21]. In addition to the mechanical challenge itself, there is increasing evidence that the enhanced passage of plasma proteins across the filtration barrier associated with glomerular hypertension may be harmful to podocytes themselves. A certain proportion of proteins passing through the filter is generally taken up by podocytes, more precisely at their basal aspect by the foot processes [153, 154]. Excessive amounts may lead to an overtaxing of the lysosomal system with the deleterious consequences generally attributed to the proximal tubule cells under these circumstances (i.e., spillage of lysosomal enzymes into the cytoplasm and increased production of reactive oxygen species, proinflammatory and profibrotic mediators). Note that in contrast to proximal tubule cells whose regenerative potential is great, podocytes are incapable of cell regeneration. Evidence that those mechanisms may play a role with respect to podocyte loss has been presented in several studies [12, 24, 96, 155]. Thus, the cells whose failure is responsible for the increased leakiness are challenged by the consequences (i.e., the increased accumulation of proteins within the filter). Consequently, proteinuria, as suggested by correlative evidence (see above), may well be a promoting factor of progression, but by damaging podocytes rather than by inducing fibrosis. In conclusion, the overload hypothesis is strongly supported by a great variety of experimental studies. In contrast, the evidence that fibrosis is responsible for progression is meager. It should be noted that the two mechanisms are not mutually exclusive, they may be additive. REGENERATION AND REPAIR OF GLOMERULAR INJURIES In recent time, the question as to the capability of a damaged nephron to undergo regeneration (restitution of structure and function) has been repeatedly raised [156– 159]. What are the chances for therapeutic interventions? Is it possible to define a point of no return, a certain degree of damage that inevitably will proceed to loss of the nephron? There are very few studies which have thoroughly addressed these questions. Therefore, we may only present some examples from individual models. They clearly show that the possibilities to repair a glomerular damage are quite limited. Again, it is useful to distinguish between endo- and extracapillary lesions. As long as an injury remains bound within the endocapillary compartment, the injury may undergo regeneration (complete restitution) or it may

414

Kriz and LeHir: Pathways to nephron loss

be repaired, leaving behind a mesangial scar (mesangial sclerosis). An extracapillary lesion associated with a tuft adhesion no longer has a chance to regenerate; the only chance for healing is the formation of an adherent scar. Endocapillary injuries The capacity of endocapillary lesions to be repaired is enormous. This is exemplified in Thy-1 nephritis, in the case of a severe endocapillary injury with an almost complete loss of capillaries and of mesangial cells, leading to the formation of a mesangial balloon. The repair followed a strict sequence [25] which, to some extent, mimics the development of a glomerular lobule during ontogeny, necessitating the orchestration of several regulatory circuits [160–168]. In this sequence in the first stage, new capillaries sprout up to sites beneath the GBM, probably attracted by podocytes. In the second stage, a-SMA– positive cells (derived from cells in the extraglomerular mesangium [169]) gather around capillaries and establish connections to the GBM in between the capillaries as well as to other centrally located mesangial cells; all together these a-SMA–positive cells form an interconnecting network fixed to the GBM and to the extraglomerular mesangial cells at the vascular pole. In the third stage, centripetal contraction of this network results in the reestablishment of GBM infoldings separating individual capillaries, thus providing a lobular texture that allows filtration. The accumulation of mesangial cells and matrix in the center of such a reestablished “lobule” represents the supporting core; it may become reduced in volume by a comparably slow remodeling process [159, 163], nevertheless, a mesangial scar (mesangial sclerosis) remains. This process is remarkable; it ensures the reestablishment of a glomerular lobule with a proper higher order structure (i.e., with peripherally arranged capillaries supported by centrally located mesangial tissue). The temporal and spatial sequence of these events appears to prevent excessive mesangial cell proliferation and matrix deposition. In less severe endocapillary damage without loss of capillaries, repair may start earlier at the second stage [25]; proliferation and subsequent contraction of a-SMA–positive cells (like in wound closure [170]), thereby reestablishing a folding pattern of the GBM, represent indispensable steps in every repair of an endocapillary injury. No data are available about whether a mesangial aneurysm with egress of blood into the mesangial space may enter a repair process and if yes when, then how this occurs. Extracapillary injuries The most characteristic feature of extracapillary lesions consists of formation of connections between the tuft and Bowman’s capsule. This includes (1) simple

Degenerative and inflammatory glomerular diseases

Endocapillary injury

Extracapillary injury

Regeneration/ repair: mesangial sclerosis

Protein leakage

Adhesions/ crescents (proteinaceous; cellular)

Excessive tubular reabsorption

Encroachment onto the tubule via glomerulotubular junction

Repair: synechia, i.e., segmental glomerulosclerosis

? Tubular atrophy

? Tubular degeneration, obstruction, and disconnection

Repair: synechia with atrophic tubular segment

Loss of viable nephrons atubular glomeruli, aglomerular tubular remnants

Tubulointerstitial inflammation/fibrosis

Overload of remaining nephrons ?

Scar

Damaging impact on healthy nephrons –––– Renal failure

Fig. 6. Flow diagram of damage progression in degenerative and inflammatory glomerular diseases (does not include collapsing glomerulopathy). A question mark adjacent to an arrow indicates the thate effect is not firmly established. The most direct path to nephron loss leads from podocyte injury to crescents, to encroachment onto the tubule, to atubular glomeruli and glomerular tubular remnants. Degeneration of these nephron remnants leads to an interstitial inflammation whose impact as a damaging factor is debated.

Kriz and LeHir: Pathways to nephron loss

cellular bridges between the GBM and the PBM (podocytes and parietal cells), (2) tuft adhesions defined in that, in addition to cells, a matrix bridge has developed, and (3) tuft adhesions associated with different kinds of crescents (note, that there are no crescents without a connecting stalk to the tuft). Even if there is no firm proof, it is plausible that, in the case that a disease process comes to an early halt, simple cell bridges may resolve. This would lead to the reestablishment of a freely floating tuft. Based on observations in human biopsies, it has been claimed that glomeruli with crescents may actually regenerate [171, 172]. It was found that the incidence of glomeruli with crescents between successive biopsies had decreased. Such data should be interpreted cautiously for the following reasons. First, the sample under analysis may not be representative for the whole kidney. Second, a fraction of crescentic glomeruli may have died and may have been replaced by a scar between samplings. Experimental data from animal models do not provide evidence that a tuft adhesion (with or without crescent), once established, may disconnect, reestablishing a freely floating tuft. A tuft adhesion may become smaller and, in random sections, it might appear that, after a certain time of recovery, there are fewer tuft adhesions. Quantification in serial sections, however, clearly show that the number of adhesions does not decrease [25]. There is agreement between observations in human material and animal studies that crescents may undergo a repair process terminating in an adherent scar. First, as well known from human biopsies and autopsies, aging crescents become fibrocellular and eventually fibrous. Breaks in the PBM allow a direct passage of cells, among them macrophages and fibroblasts from the interstitium into the crescent. Fibrosis in the crescent is probably, to a large extent, the consequence of invasion of both cell types [50, 52, 70]. There is evidence that the epithelial cells of crescents acquire some properties of fibroblasts, including the ability to synthesize collagen type I [74, 173, 174]. The repair of a crescent by fibrosis leads to an adherent scar [25, 175] (i.e., to segmental glomerulosclerosis) (Fig. 4). From animal models, we have some evidence that such a repair process consists of formation of a bridge of scar tissue that develops from both sides, from the mesangium and the interstitium, finally interconnecting the mesangium with the interstitium [25]. Taken together, our knowledge on the healing of glomerular lesions is very limited. So far, it is impossible to define a specific lesion that inevitably will proceed to the loss of the nephron. In the case that a crescent extends on the glomerulotubular junction, the chances for any repair become small. Dramatic developments like a glomerular destruction with uncontrolled bleeding into Bowman’s space as frequently encountered in rapidly progressing crescentic glomerulonephritis or disconnec-

415

tion of the tubule from its glomerulus, as described above, do not appear to have a chance to repair; these nephrons are lost forever. CONCLUSION We present a flow diagram showing the major pathways discussed in this review (Fig. 6). ACKNOWLEDGMENTS Our work underlying this review has been continuously supported by the “Deutsche Forschungsgemeinschaft,” the “Hartmann Muller¨ Stiftung,” and the “Prof. Dr. Karl und Gerhard Schiller-Stiftung.” We thank Carla Lyko for secretarial and Rolf Nonnenmacher for graphic assistance. ¨ Anatomie und ZellbiReprint requests to Wilhelm Kriz, Institut fur ologie I, Im Neuenheimer Feld 307, 69120 Heidelberg, Germany. E-mail: [email protected]

REFERENCES 1. BRUGGEMAN L, DIKMAN S, MENG C, et al: Nephropathy in human immunodeficiency virus-1 transgenic mice is due to renal transgene expression. J Clin Invest 100:84–92, 1997 2. BARISONI L, BRUGGEMAN L, MUNDEL P, et al: HIV-1 induces renal epithelial dedifferentiation in a transgenic model of HIVassociated nephropathy. Kidney Int 58:173–181, 2000 3. ROSS M, BRUGGEMAN L, WILSON PD, et al: Microcyst formation and HIV-1 gene expression occur in multiple nephron segments in HIV-associated nephropathy. J Am Soc Nephrol 12:2645–2651, 2004 4. KOPP J, KLOTMAN M, ADLER S, et al: Progressive glomerulosclerosis and enhanced renal accumulation of basement membrane components in mice transgenic for human immunodeficiency virus type 1 genes. Proc Natl Acad Sci USA 89:1577–1581, 1992 5. ROSS M, BRUGGEMAN L, HANSS B, et al: Podocan, a novel small leucine-rich repeat protein expressed in the sclerotic glomerular lesion of experimental HIV-associated nephropathy. J Biol Chem 278:33248–33255, 2003 6. RAY P, LIU X, ROBINSON L, et al: A novel HIV-1 transgenic rat model of childhood HIV-1-associated nephropathy. Kidney Int 63:2242–2253, 2003 7. EREMINA V, SOOD M, HAIGH J, et al: Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases. J Clin Invest 111:707–716, 2003 8. SMEETS B, TE LOEKE N, DIJKMAN H, et al: The parietal epithelial cell: A key player in the pathogenesis of focal segmental glomerulosclerosis in Thy-1.1 transgenic mice. J Am Soc Nephrol 15:928– 939, 2004 9. KRETZLER M, KOEPPEN-HAGEMANN I, KRIZ W: Podocyte damage is a critical step in the development of glomerulosclerosis in the uninephrectomized desoxycorticosterone rat. Virchows Arch A Pathol Anat Histopathol 425:181–193, 1994 ¨ 10. KRIZ W, HOSSER H, HAHNEL B, et al: Development of vascular pole associated glomerulosclerosis in the Fawn-hooded rat. J Am Soc Nephrol 9:381–396, 1998 11. FLOEGE J, HACKMANN B, KLIEM V, et al: Age-related glomerulosclerosis and interstital fibrosis in Milan normotensive rats: A podocyte disease. Kidney Int 51:230–243, 1997 ¨ 12. GASSLER N, ELGER M, KRANZLIN B, et al: Podocyte injury underlies the progression of focal segmental glomerulosclerosis in the fa/fa Zucker rat. Kidney Int 60:106–116, 2001 13. KIM Y, GOYAL M, KURNIT D, et al: Podocyte depletion and glomerulosclerosis have a direct relationship in the PAN-treated rat. Kidney Int 60:957–968, 2001 14. JAVAID B, OLSON J, MEYER T: Glomerular injury and tubular loss in adriamycin nephrosis. J Am Soc Nephrol 12:1391–1400, 2001

416

Kriz and LeHir: Pathways to nephron loss

15. RASCH R, NYENGAARD J, MARCUSSEN N, et al: Renal structural abnormalities following recovery from acute puromycin nephrosis. Kidney Int 62:496–506, 2002 16. NAGATA M, KRIZ W: Glomerular damage after uninephrectomy in young rats. II. Mechanical stress on podocytes as a pathway to sclerosis. Kidney Int 42:148–160, 1992 ¨ 17. NAGATA M, SCHARER K, KRIZ W: Glomerular damage after uninephrectomy in young rats. I. Hypertrophy and distortion of capillary architecture. Kidney Int 42:136–147, 1992 18. GANDHI M, OLSON J, MEYER T: Contribution of tubular injury to loss of remnant kidney function. Kidney Int 54:1157–1165, 1998 ¨ 19. KRIZ W, HAHNEL B, ROSENER S, et al: Long-term treatment of rats with FGF-2 results in focal segmental glomerulosclerosis. Kidney Int 48:1435–1450, 1995 20. GASSLER N, ELGER M, INOUE D, et al: Oligonephronia, not exuberant apoptosis, accounts for the development of glomerulosclerosis in the bcl-2 knockout mouse. Nephrol Dial Transplant 13:2509– 2518, 1998 ¨ B, et al: Angiotensin II type 1 21. HOFFMANN S, PODLICH D, HAHNEL receptor overexpression in podoytes induces glomerulosclerosis in transgenic rats. J Am Soc Nephrol 15:1475–1487, 2004 22. LE HIR M, KELLER C, ESCHMANN V, et al: Podocyte bridges between the tuft and Bowman’s capsule: An early event in experimental crescentic glomerulonephritis. J Am Soc Nephrol 12:2060–2071, 2001 23. LE HIR M, BESSE-ESCHMANN V: A novel mechanism of nephron loss in a murine model of crescentic glomerulonephritis. Kidney Int 63:591–599, 2003 24. SHIRATO I, HOSSER H, KIMURA K, et al: The development of focal segmental glomerulosclerosis in Masugi nephritis is based on progressive podocyte damage. Virchows Arch 429:255–273, 1996 ¨ 25. KRIZ W, HAHNEL B, HOSSER H, et al: Pathways to recovery and loss of nephrons in anti-Thy-1 nephritis. J Am Soc Nephrol 14:1904– 1926, 2003 26. NEUMANN I, BIRCK R, NEWMAN M, et al: SCG/Kinjoh mice: A model of ANCA-associated crescentic glomerulonephritis with immune desposits. Kidney Int 64:140–148, 2003 27. SCHWARTZ MM, LEWIS E: Focal segmental glomerular sclerosis: The cellular lesion. Kidney Int 28:968–974, 1985 28. SCHWIMMER J, MARKOWITZ G, VALERI A, et al: Collapsing glomerulopathy. Semin Nephrol 23:209–218, 2003 29. LAURINAVICIUS A, RENNKE H: Collapsing glomerulopathy—A new pattern of renal injury. Semin Diagn Pathol 19:106–115, 2002 30. SCHWARTZ MM, EVANS J, BAIN R, et al: Focal segmental glomerulosclerosis: Prognostic implications of the cellular lesion. J Am Soc Nephrol 10:1900–1907, 1999 31. D’AGATI V, FOGO A, BRUIJN J, et al: Pathologic classification of focal segmental glomerulosclerosis: A working proposal. Am J Kidney Dis 43:368–382, 2004 32. BARISONI L, KRIZ W, MUNDEL P, et al: The dysregulated podocyte phenotype: A novel concept in the pathogenesis of collapsing idiopathic focal segmental glomerulosclerosis and HIV-associated nephropathy. J Am Soc Nephrol 10:51–61, 1999 33. SCHWARTZ E, CARA A, SNOECK H, et al: Human immunodeficiency virus-1 induces loss of contact inhibition in podocytes. J Am Soc Nephrol 12:1677–1684, 2001 34. BARIETY J, NOCHY D, MANDET C, et al: Podocytes undergo phenotypic changes and express macrophagic-associated markers in idiopathic collapsing glomerulopathy. Kidney Int 53:918–925, 1998 35. BARISONI L, MOKRZYCKI M, SABLAY L, et al: Podocyte cell cycle regulation and proliferation in collapsing glomerulopathies. Kidney Int 58:137–143, 2000 36. EREMINA V, QUAGGIN S: The role of VEGF-A in glomerular development and function. Curr Opin Nephrol Hypertens 13:9–15, 2004 37. YANG Y, GUBLER M, BEAUfiLS H: Dysregulation of podocyte phenotype in idiopathic collapsing glomerulopathy and HIVassociated nephropathy. Nephron 91:416–423, 2001 38. YANG Y, JEANPIERRE C, DRESSLER G, et al: WT1 and PAX-2 podocyte expression in Denys-Drash syndrome and isolated diffuse mesangial sclerosis. Am J Pathol 154:181–192, 1999 39. KRIZ W, GRETZ N, LEMLEY K: Progression of glomerular diseases: Is the podocyte the culprit? Kidney Int 54:687–697, 1998

40. KERJASCHKI D: Caught flat-footed: podocyte damage and the molecular bases of focal glomerulosclerosis. J Clin Invest 108:1583– 1587, 2001 41. PAVENSTADT H, KRIZ W, KRETZLER M: Cell biology of the glomerular podocyte. Physiol Rev 83:253–307, 2003 42. FRIES J, SANDSTROM D, MEYER T, et al: Glomerular hypertrophy and epithelial cell injury modulate progressive glomerulosclerosis in the rat. Lab Invest 60:205–218, 1989 43. PAGTALUNAN ME, MILLER P, JUMPING-EAGLE S, et al: Podocyte loss and progressive glomerular injury in type II diabetes. J Clin Invest 99:342–348, 1997 44. MEYER T, BENNETT P, NELSON R: Podocyte number predicts longterm urinary albumin excretion in Pima Indians with type II diabetes and microalbuminuria. Diabetologia 42:1341–1344, 1999 45. LEMLEY K, LAFAYETTE R, SAFAI M, et al: Podocytopenia and disease severity in IgA nephropathy. Kidney Int 61:1475–1485, 2002 46. KRIZ W, LEMLEY K: The role of the podocyte in glomerulosclerosis. Curr Opin Nephrol Hypertens 8:489–497, 1999 47. KRIZ W, HARTMANN I, HOSSER H, et al: Tracer studies in the rat demonstrate misdirected filtration and peritubular filtrate spreading in with segmental glomerulosclerosis. J Am Soc Nephrol 12:496–506, 2001 ¨ B, et al: From segmental glomeru48. KRIZ W, HOSSER H, HAHNEL losclerosis to total nephron degeneration and interstitial fibrosis: A histopathological study in rat models and human glomerulopathies. Nephrol Dial Transplant 13:2781–2798, 1998 49. JENNETTE J: Rapidly progressive crescentic glomerulonephritis. Kidney Int 63:1164–1177, 2003 50. ATKINS R, NIKOLIC-PATERSON D, SONG Q, et al: Modulators of crescentic glomerulonephritis. J Am Soc Nephrol 7:2271–2278, 1996 51. JENNETTE J: Crescentic glomerulonephritis (chapter 14), in Heptinstall’s Pathology of the Kidney (vol 1), 5th ed., edited by Jennette J, Olson J, Schwartz MM, Silva FG, Philadelphia, New York, Lippincott-Raven Publishers, 1998, pp 625–656 52. SILVA FG, HOYER J, PIRANI C: Sequential studies of glomerular crescent formation in rats with antiglomerular basement membrane-induced glomerulonephritis and the role of coagulation factors. Lab Invest 51:404–415, 1984 53. STOUT L, SHIMAREET K, WHORTON E: Insudative lesions—Their pathogenesis and association with glomerular obsolescence in diabetes: A dynamic hypothesis based on single views of advancing human diabetic nephropathy. Hum Pathol 25:1213–1227, 1994 54. OLSON J, URDANETA AG, HEPTINSTALL R: Glomerular hyalinosis and its relation to hyperfiltration. Lab Invest 52:387, 1985 55. ANDERSON S, MEYER T, RENNKE H, et al: Control of glomerular hypertension limits glomerular injury in rats with reduced renal mass. J Clin Invest 76:612–619, 1985 56. D’AGATI V: The many masks of focal segmental glomerulosclerosis. Kidney Int 46:1223–1241, 1994 57. GROND J, KOUDSTAAL J, ELEMA J: Mesangial function and glomerular sclerosis in rats with aminonucleoside nephrosis. Kidney Int 27:405–410, 1985 58. HOWIE A, KIZAKI T, BEAMAN M, et al: Different types of segmental sclerosing glomerular lesions in six experimental models of proteinuria. J Pathol 157:141–151, 1989 59. HOWIE A, LEE S, SPARKE J: Pathogenesis of segmental glomerular changes at the tubular origin, as in the glomerular tip lesion. J Pathol 177:191–199, 1995 60. OPHASCHAROENSUK V, PIPPIN J, GORDON K, et al: Role of intrinsic renal cells versus infiltrating cells in glomerular crescent formation. Kidney Int 54:416–425, 1998 61. CATTELL V, JAMIESON S: The origin of glomrular crescents in experimental nephrotoxic serum nephritis in the rabbit. Lab Invest 39:584–590, 1978 62. CLARKE B, HAM K, TANGE JD, et al: Origin of glomerular crescents in rabbit nephrotoxic nephritis. J Pathol 139:247–258, 1983 63. HOLDSWORTH S, ALLEN D, THOMSON NM, et al: Histochemistry of glomerular cells in animal models of crescentic glomerulonephritis. Pathology 12:339–346, 1980 64. OYANAGI A, ORIKASA M, KAWACHI H, et al: Crescent-forming mechanism in an irreversible Thy-1 model in rats. Nephron 89:439– 447, 2001 65. LAN H, NIKOLIC-PATERSON D, ATKINS R: Involvement of activated

Kriz and LeHir: Pathways to nephron loss

66. 67. 68. 69. 70.

71. 72. 73.

74. 75. 76.

77.

78. 79.

80. 81. 82. 83. 84.

85. 86. 87.

periglomerular leukocytes in the rupture of Bowman’s capsule and glomerular crescent progression in experimental glomerulonephritis. Lab Invest 67:743–751, 1992 MAGIL A, WADSWORTH L: Monocyte involvement in glomerular crescents: A histochemical and ultrastructural study. Lab Invest 47:160–166, 1982 MAGIL A: Histogenesis of glomerular crescents. Immunohistochemical demonstration of cytokeratin in crescent cells. Am J Pathol 120:222–229, 1985 YOSHIOKA K, TAKEMURA T, AKANO N, et al: Cellular and noncellular compositions of crescents in human glomerulonephritis. Kidney Int 32:284–291, 1987 HANCOCK W, ATKINS R: Cellular composition of crescents in human rapidly progressive glomerulonephritis identified using monoclonal antibodies. Am J Nephrol 4:177–181, 1984 BOUCHER A, DROZ D, ADAFER E, et al: Relationship between the integrity of Bowman’s capsule and the composition of cellular crescents in human crescentic glomerulonephritis. Lab Invest 56:526– 533, 1987 JENNETTE J, HIPP C: The epithelial antigen phenotype of glomerular crescent cells. Am J Clin Pathol 86:274–280, 1986 HARRISON D, MACDONALD M: The origin of cells in the glomerular crescent investigated by the use of monoclonal antibodies. Histopathology 10:945–952, 1986 GUETTIER C, NOCHY D, JACQUOT C, et al: Immunohistochemical demonstration of parietal epithelial cells and macrophages in human proliferative extra-capillary lesions. Virchows Arch A Pathol Anat Histopathol 409:739–748, 1986 BARIETY J, HILL G, MANDET C, et al: Glomerular-epithelialmesenchymal transdifferentiation in pauci-immune crescentic glomerulonephritis. Nephrol Dial Transplant 18:1777–1784, 2003 MOELLER M, SOOfi A, HARTMANN I et al: Podocytes populate cellular crescents in a murine model of inflammatory glomerulonephritis. J Am Soc Nephrol 15:61–67, 2004 MOREL-MAROGER STRIKER L, KILLEN P, CHI E et al: The composition of glomerulosclerosis. I. Studies in focal sclerosis, crescentic glomerulonephritis, and membranoproliferative glomerulonephritis. Lab Invest 51:181–191, 1984 BERTANI T, MAZZUCCO G, MONGA G: How glomerular extracapillary proliferation might lead to loss of renal function: Light microscopic and immunohistochemical investigation. Nephron 91:74–78, 2002 REMUZZI G, BERTANI T: Pathophysiology of progressive nephropathies. N Engl J Med 339:1448–1456, 1998 MORIGI M, MACCONI D, ZOJA C, et al: Protein overload—Induced NF-kappaB activation in proximal tubular cells requires H 2 O 2 through a PKC-dependent pathway. J Am Soc Nephrol 13:1179– 1189, 2002 NATH K, FISCHEREDER M, HOSTETTER T: The role of oxidants in progressive renal injury. Kidney Int (Suppl) 45:111–115, 1994 NANGAKU M, PIPPIN J, COUSER W: C6 mediates chronic progression of tubulointerstitial damage in rats with remnant kidneys. J Am Soc Nephrol 13:928–936, 2002 SATO K, ULLRICH KJ: Serum-induced inhibition of isotonic fluid absorption by the kidney proximal tubule. II. Evidence that complement is involved. Biochim Biophys Acta 354:182–187, 1974 NANGAKU M, PIPPIN J, COUSER W: Complement membrane attack complex (C5b-9) mediates interstitial disease in experimental nephrotic syndrome. J Am Soc Nephrol 10:2323–2331, 1999 HSU S, COUSER W: Chronic progression of tubulointerstitial damage in proteinuric renal disease is mediated by complement activation: A therapeutic role for complement inhibitors? J Am Soc Nephrol 14:186–191, 2003 NATH K, HOSTETTER M, HOSTETTER T: Increased ammoniagenesis as a determinant of progressive renal injury. Am J Kidney Dis 17:654–657, 1991 CORREA-ROTTER R, HOSTETTER T, NATH K, et al: Interaction of complement and clusterin in renal injury. J Am Soc Nephrol 3:1172–1179, 1992 NATH K, HOSTETTER M, HOSTETTER T: Pathophysiology of chronic tubulointerstitial disease in rats. Interactions of dietary acid load, ammonia, and complement component C3. J Clin Invest 76:667– 675, 1985

417

88. NATH K, VERCELLOTTI G, GRANDE J, et al: Heme protein-induced chronic renal inflammation: Suppressive effect of induced heme oxygenase-1. Kidney Int 59:106–117, 2001 89. GONZALEZ-MICHACA L, FARRUGIA G, CROATT A, et al: Heme: A determinant of life and death in renal tubular epithelial cells. Am J Physiol Renal Physiol 286:F370–F377, 2004 90. EDDY A: Proteinuria and interstitial injury. Nephrol Dial Transplant 19:277–281, 2004 91. REMUZZI G: Nephropathic nature of proteinuria. Curr Opin Nephrol Hypertens 8:655–663, 1999 92. NATH K: Tubulointerstitial changes as a major determinant in the progression of renal damage. Am J Kidney Dis 20:1–17, 1992 93. BURTON C, HARRIS K: The role of proteinuria in the progression of chronic renal failure. Am J Kidney Dis 27:765–775, 1996 94. BENIGNI A, REMUZZI G: How renal cytokines and growth factors contribute to renal disease progression. Am J Kidney Dis 37:S21– S24, 2001 95. THOMAS ME, HARRIS K, WALLS J, et al: Fatty acids exacerbate tubulointerstitial imjury in protein-overload porteinuria. Am J Physiol Renal Physiol 283:F640–F647, 2002 96. ABBATE M, ZOJA C, MORIGI M, et al: Tranforming growth factor-b1 is up-regulated by podocytes in response to excess intraglomerular passage of proteins. Am J Pathol 161:2179–2193, 2002 97. STRUTZ F, ZEISBERG M, ZIYADEH FN, et al: Role of basic fibroblast growth factor-2 in epithelial-mesenchymal transformation. Kidney Int 61:1714–1728, 2002 98. PAVENSTADT H: Roles of the podocyte in glomerular function. Am J Physiol 278:F173–F179, 2000 99. KEES-FOLTS D, SADOW J, SCHREINER GF: Tubular catabolism of albumin is associated with the release of an inflammatory lipid. Kidney Int 45:1697–1709, 1994 100. KANG D, KANELLIS J, HUGO C, et al: Role of the microvascular endothelium in progressive renal disease. J Am Soc Nephrol 13:806– 816, 2002 101. NAJAfiAN B, KIM Y, CROSSON J, et al: Atubular glomeruli and glomerulotubular juction abnormalities in diabetic nephropathy. J Am Soc Nephrol 14:908–917, 2003 102. ASSMANN K, VAN SON JPHF, DIJKMAN H, et al: Antibody-induced albuminuria and accelerated focal glomerulosclerosis in the Thy1.1 transgenic mouse. Kidney Int 62:116–126, 2002 103. MARCUSSEN N: Biology of disease: Atubular glomeruli and structural basis for chronic renal failure. Lab Invest 66:265–284, 1992 104. MARCUSSEN N, CHRISTENSEN S, PETERSEN J, et al: Atubular glomeruli, renal function and hypertrophic response in rats with chronic lithium nephropathy. Virchows Arch 419:281–289, 1991 105. MARCUSSEN N: Tubulointerstitial damage leads to atubular glomeruli: Significance and possible role in progression. Nephrol Dial Transplant 15:74–75, 2000 106. MARCUSSEN N: Atubular glomeruli in chronic renal disease. Curr Top Pathol 88:145–174, 1995 107. COHEN E, REGNER K, FISH B, et al: Stenotic glomerulotubular necks in radiation nephropathy. J Pathol 190:484–488, 2000 ¨ H, WARNECKE E, OLBRICHT C: Characteristics of renal tubu108. GRONE lar atrophy in experimental renovascular hypertension: A model of kidney hibernation. Nephron 72:243–252, 1996 109. WILLIAMS JD, COLES G: Proteinuria—A direct cause of renal morbidity? Kidney Int 45:443–450, 1994 110. ONG ACM, FINE L: Tubular-derived growth factors and cytokines in the pathogenesis of tubulointerstitial fibrosis: Implications for human renal disease progression. Am J Kidney Dis 23:205–209, 1994 111. ZOJA C, MORIGI M, REMUZZI G: Proteinuria and phenotypic change of proximal tubular cells. J Am Soc Nephrol 14:S36–S41, 2003 112. MEYER T: Tubular injury in glomerular disease. Kidney Int 63:774– 787, 2003 113. TRYGGVASON K, PETTERSSON E: Causes and consequences of proteinuria: The kidney filtration barrier and progressive renal failure. J Intern Med 254:216–224, 2003 114. STRUTZ F, NEILSON E: New insight into mechanisms of fibrosis in immune renal injury. Springer Semin Immunpathol 24:459–476, 2003

418

Kriz and LeHir: Pathways to nephron loss

115. D’AMICO G, FERRARIO F, RASTALDI M: Tubulointerstitial damage in glomerular diseases: Its role in the progression of renal damage. Am J Kidney Dis 26:124–132, 1995 116. CAMERON J: Tubular and interstitial factors in the progression of glomerulonephritis. Pediatr Nephrol 6:292–303, 1992 117. BOHLE A, GLOMB D, GRUND K, et al: Correlations between relative interstitial volume of the renal cortex and serum creatinine concentration in minimal changes with nephrotic syndrome and in focal sclerosing glomerulonephritis. Virchows Arch A Pathol Anat Histopathol 376:221–232, 1977 ¨ O, et al: Long-term prog118. WEHRMANN M, BOHLE A, BOGENSCHUTZ nosis of chronic idiopathic membranous glomerulonephritis. An analysis of 334 cases with particular regard to tubulointerstitial changes. Clin Nephrol 31:67–76, 1989 119. SCHAINUCK LI, STRIKER G, CUTLER R, et al: Structural-functional correlations in renal disease. Part II: The correlations. Hum Pathol 1:631–641, 1970 120. RISDON R, SLOPER JC, DE WARDENER H: Relationship between renal function and histologic changes found in renal-biopsy specimens from patients with persistent glomerular nephritis. Lancet 2:363–366, 1968 121. D’AMICO G, BAZZI C: Pathophysiology of proteinuria. Kidney Int 63:809–825, 2003 122. TANG S, LEUNG J, ABE K, et al: Albumin stimulates interleukin-8 expresison in proximal tubular epithelial cells in vitro and in vivo. J Clin Invest 111:515–527, 2003 123. NOCHY D, HEUDES D, GLOTZ D, et al: Preeclampsia associated focal and segmental glomerulosclerosis and glomerular hypertrophy: A morphometric analysis. Clin Nephrol 42:9–17, 1994 124. EDDY A, MCCULLOCH L, ADAMS J, et al: Interstitial nephritis induced by protein-overload proteinuria. Am J Pathol 135:718–719, 1989 125. SHIMIZU H, MARUJAMA S, YUZAWA Y, et al: Anti-monocyte chemoattractant protein-1 gene therapy attenuates renal injury induced by protein-overload proteinuria. J Am Soc Nephrol 14:1496– 505, 2003 126. KIKUCHI H, KAWACHI H, ITO Y, et al: Severe proteinuria, sustained for 6 months, induces tubular epithelial cell injury and cell infiltration in rats but not progressive interstitial fibrosis. Nephrol Dial Transplant 15:799–810, 2000 127. OKUDA S, OOCHI N, WAKISAKA M, et al: Albuminuria is not an aggravating factor in experimental focal glomerulosclerosis and hyalinosis. J Lab Clin Med 119:245–253, 1992 128. LEMLEY K, KRIZ W: Glomerular injury in analbuminemic rats after subtotal nephrectomy. Nephron 59:104–109, 1991 129. REMUZZI G, RUGGENENTI P, BENIGNI A: Understanding the nature of renal disease progression. Kidney Int 51:2–15, 1997 130. NYENGAARD J, BENDTSEN T: Glomerular number and size in relation to age, kidney weight, and body surface in normal man. Anat Rec 232:194–201, 1992 131. LUBRAN M: Renal function in the elderly. Ann Clin Lab Sci 25:122– 133, 1995 132. BAJEMA I, HAGEN E, HERMANS J, et al: Kidney biopsy as a predictor for renal outcome in ANCA-associated necrotizing glomerulonephritis. Kidney Int 56:1751–1758, 1999 133. BRENNER B, MEYER T, HOSTETTER T: Dietary protein intake and the progressive nature of kidney disease: The role of hemodynamically mediated glomerular injury in the pathogenesis of progressive glomerular sclerosis in aging, renal ablation, and intrinsic renal disease. N Engl J Med 307:652–659, 1982 134. HOSTETTER T, OLSON J, RENNKE H, et al: Hyperfiltration in remnant nephrons: A potentially adverse response to renal ablation. Am J Physiol 241:F85–F93, 1981 135. FOGO A, HAWKINS E, BERRY P, et al: Glomerular hypertrophy in minimal change disease predicts subsequent progression to focal glomerular sclerosis. Kidney Int 38:115–123, 1990 136. SIMONS JL, PROVOOST A, ANDERSON S, et al: Modulation of glomerular hypertension defines susceptibility to progressive glomerular injury. Kidney Int 46:396–404, 1994 137. RENNKE H, KLEIN P: Pathogenesis and significance of nonprimary focal and segmental glomerusclerosis. Am J Kidney Dis 13:443– 456, 1989 138. SHEA S, RASKOVA J, MORRISON A: A stereologic study of glomeru-

139. 140. 141. 142. 143. 144. 145. 146.

147. 148. 149.

150.

151. 152. 153. 154. 155. 156.

157. 158. 159. 160.

161.

162.

lar hypertrophy in the subtotally nephrectomized. Am J Pathol 90:201–210, 1978 MEYER T, RENNKE H: Progressive glomerular injury after limited renal infarction in the rat. Am J Physiol 254:F856–F862, 1988 MEYER T, ANDERSON S, RENNKE H, et al: Reversing glomerular hypertension stabilizes established glomerular injury. Kidney Int 31:752–759, 1987 MACKAY K, STRIKER L, STAUFFER JW, et al: Relationship of glomerular hypertrophy and sclerosis: Studies in SV40 transgenic mice. Kidney Int 37:741–748, 1990 MILLER P, RENNKE H, MEYER T: Glomerular hypertrophy accelerates hypertensive glomerular injury in rats. Am J Physiol 261:F459– F465, 1991 DWORKIN L, BERNSTEIN J, TOLBERT E, et al: Salt restriction inhibits renal growth and stabilizes injury in rats with established renal disease. J Am Soc Nephrol 7:437–442, 1996 HOSTETTER T: Hyperfiltration and glomerulosclerosis. Semin Nephrol 23:194–199, 2003 HILL G, HEPTINSTALL R: Steroid-induced hypertension in the rat. Am J Pathol 51:1–39, 1968 VERSEPUT GH, PROVOOST A, BRAAM B, et al: Angiotensinconverting enzyme inhibition in the prevention and treatment of chronic renal damage in the hypertensive fawn-hooded rat. J Am Soc Nephrol 8:249–259, 1997 IVERSEN B, AMANN K, KVAM F, et al: Increased glomerular capillary pressure and size mediate glomerulosclerosis in SHR juxtamedullary cortex. Am J Physiol 274:F365–F373, 1998 ENDLICH N, KRESS K, REISER J, et al: Podocytes respond to mechanical stress in vitro. J Am Soc Nephrol 12:413–422, 2001 ENDLICH N, SUNOHARA M, NIETfiELD W, et al: Analysis of differential gene expression in stretched podocytes: osteopontin enhances adaptation of podocytes to mechanical stress. FASEB J 16:1850– 1852, 2002 MORTON MJ, HUTCHINSON K, MATHIESON PW, et al: Human podocytes possess a stretch-sensitive, Ca2+-activated K+ channel: Potential implications for the control of glomerular filtration. J Am Soc Nephrol 15:2981–2987, 2004 SHARMA M, SHARMA R, GREENE A, et al: Documentation of angiotensin II receptors in glomerular epithelial cells. Am J Physiol 274:F623–F627, 1998 DURVASULA R, PETERMANN A, HIROMURA K, et al: Activation of a local tissue angiotensin system in podocytes by mechanical strain. Kidney Int 65:30–39, 2004 KERJASCHKI D, EXNER M, ULLRICH R, et al: Pathogenic antibodies inhibit the binding of apolipoproteins to megalin/gp330 in passive Heymann nephritis. J Clin Invest 100:2303–2309, 1997 KERJASCHKI D: Megalin/GP330 and pathogenetic concepts of membranous glomerulopathy (MGN). Kidney Blood Press Res 23:163– 166, 2000 JOLES J, KUNTER U, JANSSEN U, et al: Early mechanisms of renal injury in hypercholesterolemic or hypertriglyceridemic rats. J Am Soc Nephrol 11:669–683, 2000 HAUER H, BAJEMA I, VAN HOUWELINGEN H, et al: Determinants of outcome in ANCA-associated glomerulonephritis: A prospective clinico-histopathological analysis of 96 patients. Kidney Int 62:1732–1742, 2002 FOGO A: The potential for regression of renal scarring. Curr Opin Nephrol Hypertens 12:223–225, 2003 MA L, NAKAMURA S, WHITSITT J, et al: Regression of sclerosis in aging by an angiotensin inhibition-induced decrease in PAI-1. Kidney Int 58:2425–2436, 2000 SAVILL J, JOHNSON R: Glomerular remodelling after inflammatory injury. Exp Nephrol 3:149–158, 1995 MASUDA Y, SHIMIZU A, MORI T, et al: Vascular endothelial growth factor enhances glomerular capillary repair and accelerates resolution of experimentally induced glomerulonephritis. Am J Pathol 159:599–608, 2001 OSTENDORF T, KUNTER U, VAN ROEYEN C, et al: The effects of plateled-derived growth factor antagonism in experimental glomerulonephritis are independent of the transforming growth factor-beta system. J Am Soc Nephrol 13:658–667, 2002 IRUELA-ARISPE L, GORDON K, HUGO C, et al: Participation of

Kriz and LeHir: Pathways to nephron loss

163.

164. 165. 166. 167. 168. 169.

glomerular endothelial cells in the capillary repair of glomerulonephritis. Am J Pathol 147:1715–1727, 1995 BAKER A, MOONEY A, HUGHES J, et al: Mesangial cell apoptosis: The major mechanism for resolution of glomerular hypercellularity in experimental mesangial proliferative glomerulonephritis. J Clin Invest 94:2105–2116, 1994 ROBERT B, ABRAHAMSON D: Control of glomerular capillary development by growth factor/receptor kinases. Pediatr Nephrol 16:294– 301, 2001 ITO Y, GOLDSCHMEDING R, BENDE R, et al: Kinetics of connective tissue growth factor expression during experimental proliferative glomerulonephritis. J Am Soc Nephrol 12:472–484, 2001 ¨ H, et al: Specific antogonism of OSTENDORF T, KUNTER U, GRONE PDGF prevents renal scarring in experimental glomerulonephritis. J Am Soc Nephrol 12:909–918, 2001 BLOM I, VAN DIJK A, WIETEN L, et al: In vitro evidence for differential involvement of CTGF, TGFbeta, and PDGF-BB in mesangial response to injury. Nephrol Dial Transplant 16:1139–1148, 2001 CHEN Y, BLOM I, SA S, et al: CTGF expression in mesangial cells: involvement of SMADs, MAP kinase, and PKC. Kidney Int 62:1149– 1159, 2002 HUGO C, SHANKLAND SJ, BOWEN-POPE D, et al: Extraglomerular

170. 171. 172.

173.

174. 175.

419

origin of the mesangial cell after injury- a new role of the juxtaglomerular apparatus. J Clin Invest 100:786–794, 1997 NEDELEC B, GHAHARY A, SCOTT PG, et al: Control of wound contraction. Basic and clinical features. Hand Clin 16:289–302, 2000 HAUER H, BAJEMA I, HAGEN E, et al: Long-term renal injury in ANCA-associated vasculitis: an analysis of 31 patients with followup biopsies. Nephrol Dial Transplant 17:587–596, 2002 BAGCHUS W, HOEDEMAEKER P, ROZING J, et al: Glomerulonephritis induced by monoclonal anti-Thy 1.1 antibodies. A sequential histological and ultrastructural study in the rat. Lab Invest 55:680–687, 1986 HE J, HAYASHI K, HORIKOSHI S, et al: Identification of cellular origin of type I collagen in glomeruli of rats with crescentic glomerulonephritis induced by anti-glomerular basement membrane. Nephrol Dial Transplant 16:704–711, 2001 NG Y, FAN J, MU W, et al: Glomerular epithelial-myofibroblast transdifferentiation in the evolution of glomerular crescent formation. Nephrol Dial Transplant 14:2860–2872, 1999 HOWIE A, FERREIRA M, MAJUMDAR A, et al: Glomerular prolapse as precursor of one type of segmental sclerosing lesions. J Pathol 190:478–493, 2000