Renal epithelial injury and fibrosis

Renal epithelial injury and fibrosis

    Renal epithelial injury and fibrosis Brigitte Kaissling, Michel LeHir, Wilhelm Kriz PII: DOI: Reference: S0925-4439(13)00055-0 doi: ...

1MB Sizes 0 Downloads 99 Views

    Renal epithelial injury and fibrosis Brigitte Kaissling, Michel LeHir, Wilhelm Kriz PII: DOI: Reference:

S0925-4439(13)00055-0 doi: 10.1016/j.bbadis.2013.02.010 BBADIS 63643

To appear in:

BBA - Molecular Basis of Disease

Received date: Revised date: Accepted date:

26 November 2012 7 February 2013 9 February 2013

Please cite this article as: Brigitte Kaissling, Michel LeHir, Wilhelm Kriz, Renal epithelial injury and fibrosis, BBA - Molecular Basis of Disease (2013), doi: 10.1016/j.bbadis.2013.02.010

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

IP

SC R

Renal epithelial injury and fibrosis

T

Special issue of "BBA-Molecular Basis of Disease (ELSP)"

Brigitte Kaissling1, Michel LeHir1 and Wilhelm Kriz2

NU

1 Institute of Anatomy, University of Zurich, Zurich, Switzerland

TE

D

MA

2 Institute of Transfusion Medicine and Immunology, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany

AC

CE P

To whom correspondence should be directed: Wilhelm Kriz, MD 
 D 68167 Mannheim, Ludolf-Krehl-Str.13-17, Germany Phone: ++49 621 383 9941 Fax: ++49 621 383 9949 [email protected]

1

ACCEPTED MANUSCRIPT ROAD MAP Introduction and question

IP

T

Fibrosis development

SC R

Fibrosis development starting from glomerular diseases

NU

Tubular degeneration subsequent to glomerular injuries

MA

Peritubular response to tubule degeneration: the myofibroblast and fibrosis development

D

Fibrosis development starting from tubular injuries

TE

Signalling underlying the tubulo-interstitial phenomena

CE P

Relevance of fibrosis for progression of CKD?

AC

Relevance of fibrosis for the progression of CKD subsequent to glomerular diseases Relevance of fibrosis in the progression of a tubular injury to CKD Concluding Remarks

2

ACCEPTED MANUSCRIPT

Abstract

T

Chronic kidney disease at a certain advanced stage inevitably

IP

progresses to end stage renal failure characterized by the

SC R

progressing loss of nephrons accompanied by the increasing appearance of fibrotic tissue, called renal fibrosis. The urgent question is whether renal fibrosis is a response to injury or if

NU

fibrosis acquires a self-sustaining progressive potential that

MA

actively contributes to the deterioration of the kidney.

The present review distinguishes between renal fibrosis

TE

D

subsequent to a glomerular injury and fibrosis subsequent to a primary tubular injury. Glomerular diseases enter a progressing

CE P

course after encroaching onto the tubule leading to what is generally called "tubulointerstitial fibrosis". The progression of

AC

the injury at the level of the tubulointerstitium appears to be fully dependent on the progression of the disease in the corresponding glomerulus

Primary tubular injuries have a very good chance of recovery. If they develop a local fibrotic process, this seems to be supportive for recovery. Cases in which recovery fails appear to secondarily initiate a glomerular disease accounting for a glomerulus-dependent vicious cycle to progression. Even if most researchers think of renal fibrosis as a process promoting the progression of the disease this review points out 3

ACCEPTED MANUSCRIPT that the available structural evidence speaks in favour of a protective role of fibrosis supporting recovery after acute tubular

T

injury or, under progressing circumstances, providing a firm

IP

three-dimensional framework that permits still intact or partially

SC R

damaged nephrons to survive.

NU

Keywords: tubulointerstitial fibrosis-tubular injury-glomerular injury-progression to CKD

MA

Highlights

TE

D

Fibrosis associated with a tubular injury is supportive for recovery

CE P

Fibrosis associated with a glomerular disease fully depends on the progression of the glomerular injury.

AC

Fibrosis provides secured spaces for intact or partially injured nephrons to survive Fibrosis is no promoter of progression, in contrast, fibrosis is protective

4

ACCEPTED MANUSCRIPT

Introduction and question

T

The present review deals with mechanisms of renal fibrosis

IP

development secondary to an epithelial injury, i.e. secondary to

SC R

an injury to the glomerulus or to the tubule. Renal fibrosis derived from an interstitial or vascular disease is excluded. The

NU

present review predominantly relies on structural observations

MA

in animal models.

Nephrons degenerate individually and largely independent from

D

neighbouring nephrons [1]. In case that a nephron undergoes

TE

complete degeneration a considerable micro volume of an

CE P

estimated 40x106 m3 (rat) of epithelial tissue is lost. That volume will be contracted and filled by fibrous tissue. Since most of this volume is due to the loss of the tubule (an

AC

estimated 95% of a nephron is made up by the tubule), renal fibrosis is generally understood as "tubulointerstitial fibrosis", as a process that occurs in conjunction with tubule degeneration [2-4]. Glomerular fibrosis, i.e. glomerulosclerosis, is considered as an own disease entity, though, by inducing tubular degeneration, it represents the cause of tubulointerstitial fibrosis in the majority of cases progressing to end stage renal disease [1].

5

ACCEPTED MANUSCRIPT The term "tubulointerstitial fibrosis" denominates the pathology with extensive scar formation encountered in autopsies/biopsies

T

of end stage kidneys as well as all intermediate stages of

IP

fibrosis development, including early stages of focal nephron

SC R

degeneration/atrophy locally associated with interstitial matrix production. Descriptively this term correctly characterizes the pathology, mechanistically, however, it is a biased term. Since it

NU

points to fibrosis instead of tubular injury as the central

MA

condition, it "imputes a degree of autonomy to fibrosis that neglects the tubular component" [5].

TE

D

The crucial question, whether the processes in the tubule or those in the interstitium are the primary driving force for

CE P

progression of chronic kidney disease (CKD) remains a matter of intense discussion. In general it is simply assumed, without

AC

explicit justification, that fibrosis is harmful. According to that assumption, foci of "tubulointerstitial fibrosis", i.e. ongoing local inflammatory processes associated with injured nephrons, will not only accelerate the degeneration of those nephrons but they may also encroach onto neighbouring healthy tubules, thereby becoming an active contributor to progression. The present review will propose another point of view, whereby fibrosis develops secondary to tubular injury. In this concept fibrosis either participates as a healing process or it produces a framework of dense fibrotic tissue that supports the survival of healthy and partially damaged nephrons. 6

ACCEPTED MANUSCRIPT

T

Fibrosis development

IP

Diseases of the nephron that progress to fibrosis have to be

SC R

subdivided into those beginning in the glomerulus and those beginning in the tubule. Glomerular diseases make up the overwhelming majority of diseases, which progress to CKD.

NU

Primary tubular diseases generally display as acute kidney

MA

injury (AKI or acute renal failure), in which the majority of affected tubules undergoes complete regeneration [6]. However, in a small fraction the regenerating processes may

TE

D

enter a prolonged, a chronic development also showing up as

CE P

foci of "tubulointerstitial fibrosis

AC

Fibrosis development in diseases beginning in the glomerulus

A large number of inflammatory and degenerative glomerular diseases are prone to progress to CKD displaying different histopathological features and time courses of progression. It is of utmost importance to understand that, although many nephrons may be affected at one time, the evolution of the disease is specific for each individual nephron and solely dependant on the evolution in the respective glomerulus; the progression of the disease process in the tubule is secondary to that in the glomerulus.

7

ACCEPTED MANUSCRIPT Progression of a glomerular disease starts with the loss of the separation between the tuft and Bowman's capsule, i.e. the

T

formation of an adhesion of the tuft to Bowman's capsule. In

IP

degenerative conditions this process leads to focal segmental

SC R

glomerulo-sclerosis (FSGS), in inflammatory conditions to crescentic glomerulonephritis. In both, the outcome with respect to progression is very similar [1]. Both processes start as injury

NU

of a glomerular segment and both progress within Bowman's

MA

capsule, i.e. between the parietal epithelium and the parietal basement membrane. Both processes may encroach on other lobules (eventually leading to global process) and/or on the

TE

D

glomerulo-tubular junction, leading to a narrowing and finally to a complete obstruction of the urinary orifice (Fig.1). The

CE P

associated decrease of filtrate delivery, finally depriving the tubule from any workload, causes atrophy, finally degeneration

lost.

AC

of the corresponding tubule. In the end the entire nephron is

This process may favourably be studied in mouse or rat models of rapidly progressing glomerular diseases (e.g. DOCA-salt hypertension in rats [7]; THY-1 mediated glomerulonephritis [8] and anti-GBM nephritis in mice [9]). In these models, the degenerative process advances undelayed to full nephron degeneration and focal fibrosis. Comparing the advancement of the damage in individual nephrons which are caught in various 8

ACCEPTED MANUSCRIPT stages of disease provides a reliable basis for the

T

understanding of the sequence of damage progression.

IP

In addition to the encroachment hypothesis two other

SC R

mechanisms have been proposed to account for the transfer of the glomerular injury onto the tubule and the interstitium. First, the leakage of substances (proteins) through the damaged

NU

glomerular filter that exert toxic effects on the cells of the

MA

proximal tubule [10] leading to the development of fibrosis. However, recent studies [11,12] have shown that protein leakage induces proliferation of tubule cells, not degeneration.

TE

D

Second, it has been proposed, that sclerosis and loss of postglomerular microvasculature related to the affected

CE P

glomerulus cause decreased peritubular perfusion of the corresponding tubule [13,14]. That hypothesis overlooks the

AC

fact, that the peritubular capillary network is commonly supplied by all glomeruli in the area [15]. Moreover, a recent study in von Hippel Lindau-knockout mice did not found supportive evidence for this hypothesis [16].

Tubular degeneration subsequent to glomerular injuries

A straightforward course of tubular degeneration is consistently seen after complete obstruction of the urinary orifice by a sclerotic or crescentic lesion [1] The deprivation of the tubule from any filtrate delivery leads to inactivity atrophy rapidly 9

ACCEPTED MANUSCRIPT progressing to complete decomposition of the tubule. The degenerative process starts in the most proximal tubule

T

segments just behind the glomerulo-tubular junction and

IP

proceeds distally (Fig.2A). Tubules are cylindrical structures

SC R

consisting of a single epithelial cell layer resting on the tubular basement membrane (TBM). During degeneration of a tubule the epithelial cells degenerate and disappear well before the

NU

dissolution of the TBM. Thus until their death, tubular cells

MA

remain separated from the interstitium by the TBM. The removal of the TBM occurs much later. As presented in detail in recent work in the UUO model [17] and in an injury model in which

TE

D

overproduction of TGFb1 caused cellular demise [18] autophagy represents the major mechanism of tubular epithelial

CE P

cell death (Fig.2B), fratricide may participate. In response to direct toxic effects on tubular cells other mechanisms of cell

AC

death (necrosis, apoptosis) may prevail [19]. The decomposition and removal of the cells may finally lead to almost cell-free cylinders of the TBM, which become increasingly wrinkled and eventually collapsed.

The TBM maintains its continuity for considerable time even after collapsing and after disappearance of any cellular elements inside these cylinders. Tubular cells crossing the TBM border of degenerating tubules were not encountered in our studies nor documented by others. Even in case of a complete degeneration of a tubule, removal of its remnants and formation 10

ACCEPTED MANUSCRIPT of a scar, the glomerular remnants may survive for some time

T

as inoperable structures [8,20,21]

IP

Extended segmental sclerosis without encroachment on the

SC R

glomerulo-tubular junction, or an incomplete obstruction of the urinary orifice will result in a significant decrease of filtration and the workload of the tubule will be markedly reduced. Under

NU

those conditions tubular cells do not completely degenerate but

MA

undergo atrophy with loss of features characteristic of a differentiated phenotype [22]. Such intermediate stages of tubule degeneration appear to cause very similar, if not identical

TE

D

responses of the surrounding interstitium - like they are seen after straightforward degeneration of the tubule (see below).

CE P

They may obviously persist over extended periods of time in humans. The rapid progression in most animal models does not

AC

provide the opportunity to observe that type of lesions. Therefore details of their evolution are unknown.

Peritubular response to tubule degeneration: myofibroblasts and fibrosis development

Tubule degeneration evokes a vivid peritubular reaction. The injured tubule cells initiate a local inflammation, which will finally lead to the removal of all tubular remnants and the formation of a local area of fibrosis, finally a scar. The outstanding feature of the local inflammatory process surrounding a degenerating 11

ACCEPTED MANUSCRIPT tubule consists of the massive accumulation of myofibroblasts. The myofibroblast is a central player in fibrosis development.

T

According to our results, the only relevant source for these

IP

myofibroblasts are the resident interstitial cells [23,24] (Fig.3),

SC R

which are generally called interstitial fibroblasts, some authors call them perivascular fibroblasts or pericytes [25]. The much discussed proposal that myofibroblasts may also develop from

NU

injured tubular cells by a process called "Epithelial to

MA

Mesenchymal Transition (EMT)" has not found supporting evidence in recent studies [18,23,25]. Also a contribution of bone-marrow derived "fibrocytes" is generally considered as

TE

D

marginal, at best [26]. Myofibroblasts organize the interstitial fibrosis by synthesizing type I collagen and finally by contraction

AC

CE P

of the affected area to a scar.

Thus, the differentiation of resident interstitial fibroblasts into myofibroblasts is central in fibrosis development. Interstitial fibroblasts are branched cells with partially very long cellular projections that are stuffed with bundles of actin filaments. The processes are attached (by a kind of adherens junctions) to the basement membrane of tubules and peritubular capillaries as well as to processes and cell bodies of other fibroblasts [27], forming a continuous network throughout the interstitium (Fig.3).

12

ACCEPTED MANUSCRIPT Upon activation by profibrotic cytokines (see below) or mechanical stress [28] resident interstitial fibroblasts

T

progressively gain the myofibroblast phenotype as consistently

IP

seen in various disease models including the UUO model [24].

SC R

Within one day after UUO the interstitial fibroblasts started to proliferate in the vicinity of injured tubules [24], alpha smooth muscle actin (αSMA), the hallmark of myofibroblasts, became

NU

detectable by immunofluorescence in interstitial fibroblasts and

MA

after two days bundles of alpha-SMA traversed the cells in all directions and accumulated in their processes. The contacts to neighbouring fibroblasts / myofibroblast increased in number,

TE

D

the area of contacts to injured tubules broadened, partially encircling a tubule (Fig. 2B). Thus, the network of fibroblasts

CE P

developed into a network of myofibroblasts (Fig.4) suggesting that this conversion is not simply a response of individual cells

AC

but of a cellular network. The cells produce abundant amounts of matrix and collagen fibrils that fill the meshes of this network. In total this newly emerged fibrous tissue forms an area with strong mechanical properties centripetally interconnecting all structures of this area

In case of complete tubular degeneration the tubular remnants finally disappeared and the girdles of fibrotic tissue formed around the degenerating tubules were combined to areas of scar tissue as routinely seen in biopsies/autopsies of human endstage kidneys. In case that the degeneration of the tubule is 13

ACCEPTED MANUSCRIPT prolonged with maintenance of atrophic tubules for some time, the peritubular inflammatory response persists as well.

T

Unfortunately, such advanced stages of fibrosis development

SC R

IP

have never been studied in detail in an animal model.

Fibrosis development starting from tubular injuries

NU

Acute kidney injury (AKI) as clinical disease most frequently

MA

occurs in conjunction with multiple organ failure. Many factors (hemodynamic, inflammatory, endothelial, toxic, metabolic) may converge to cause the tubular injury. Here, we will only review

TE

D

the disease mechanisms subsequent to primary tubular injuries as they are encountered in response to direct hypoxic (ischemic

CE P

reperfusion injury) or toxic/metabolic challenges and as they

AC

have been studied in animal models.

Under such experimental conditions the entire population of nephrons is affected, however, the severity of the damage may be very different among nephrons. The immediate loss of a nephron occurs only in the rare cases that the tubular basement membrane is disrupted (tubulorrhexis) [29]. Generally, even after severe insults the TBM maintains its continuity providing the scaffold for epithelial regeneration.

The epithelial damage is generally heterogeneous among segments within a nephron and among cells within a segment 14

ACCEPTED MANUSCRIPT of the nephron. Some cells are lethally injured and undergo necrosis or apoptosis (the cell debris seems to be washed away

T

by urine flow) [30,31] others survive the attack with various

IP

degrees of injury. The latter start to proliferate vigorously,

SC R

replacing lost cells and finally differentiate into functioning tubular cells re-establishing a functional tubule in most cases.

NU

Ischemic events after ligation of the renal artery and some toxic

MA

events (e.g. uranyl acetate [32], cisplatin [33], mercury chloride (HgCl2) [34] predominantly attack the S3 segments of the proximal tubules. Distal tubular segments are targeted by

TE

D

agents that interfere with the segment-specific function (Fig.5). For instance, uromodulin-associated kidney disease [35] affects

CE P

specifically the thick ascending limb of Henle’s loop (TAL), the site of uromodulin synthesis, thiazide treatment in rats [30,31]

AC

affects exclusively the distal convoluted tubule (DCT; Fig. 5), the location of the thiazide-sensitive neutral sodium cotranporter (NCC) and distal tubules and collecting ducts are injured by lithium [36] which interferes with the function of these segments.

With respect to recovery of a tubular injury the question arises whether there is a certain fraction of cells capable of regeneration and another fraction that is not. Considering the high degree of structural homogeneity among cells in the proximal tubule, the TAL and DCT it is unlikely that two separate cell populations might co-exist, with different capacity 15

ACCEPTED MANUSCRIPT to survive insults and to proliferate. The heterogeneity in cell fate observed after injury likely results from differences in the

T

cycling stage of individual cells. Differentiated tubular cells are

IP

cycling at a slow rate, even in the adult [37,38]. As shown

SC R

previously the proximal tubule contains a large fraction of cells arrested in G1 that constitutes a "ready-to-go" pool of cells to immediately start proliferation in case of a tubular injury [38]. It

NU

was also shown [39] that proximal tubular injury by HgCl2

MA

induced expression of the nuclear factor NFATc1 in some cells, which thereby became more resistant to cell death (apoptosis). Fate tracing showed that those cells were largely responsible

TE

D

for the regeneration of the tubular epithelium. Thus, there seem to be relevant differences between tubular cells concerning their

CE P

stage in the cell cycle and in their vulnerability to toxic or ischemic challenges resulting in a fraction of cells being prone

AC

for regenerative proliferation. A very recent study in rats uncovered the emergence of a population of CD24, CD133 and vimentin positive cells in proximal tubules undergoing regeneration [40] confirming previous studies showing that the expression of vimentin characterizes a fraction of regenerating proximal tubule cells [34] On the other hand, two recent studies [5,6,41] show that there is or develops a fraction of cells that does not properly participate in the process of regeneration. These cells are arrested in G2/M delaying proliferation and produce JNK (c-jun NH2-terminal kinase) dependent profibrotic factors including 16

ACCEPTED MANUSCRIPT connective tissue growth factor (CTGF) and transforming growth factor beta (TGFß) [41]. In the other study such cells

T

were shown to be characterized by deficiency or loss of

IP

phosphatase and tensin homolog (PTEN) likely evoked by

SC R

TGFß in autocrine or paracrine fashion [5]. They are likely the cause of the failure of some tubules to regenerate,

NU

consequently the triggers of the peritubular inflammation.

MA

In summary, injured tubules seem to contain a fraction of cells, dependant on their stage in the cell cycle that accounts for the great regeneration capacity of tubular epithelia whereas other

TE

D

cells do not have these outstanding capabilities for proliferation

CE P

and redifferentiation.

In case that tubules fail to recover, a signalling activity is

AC

maintained that evokes a local peritubular inflammation. Interstitial fibroblast start to proliferate and to convert into myofibroblasts followed by matrix deposition leading to foci of "tubulointerstitial fibrosis". When comparing figures 2B and 5B it is evident that the inflammatory response to a primary tubular injury is structurally identical to that associated with a glomerular-dependant tubular injury.

In a series of elegant studies in the uranyl acetate-induced AKI in rats of the Akira Hishida group [32,42-44] it was shown that the interstitial response is spatially very precisely correlated to 17

ACCEPTED MANUSCRIPT the affected tubular segment and cells, and disappears precisely at loci where the tubular cells recovered. This strongly

T

suggests that the injured tubular cells in small groups or even

IP

individually are responsible for the inflammation and fibrosis in

SC R

their immediate surroundings. Moreover, the authors did not find any evidence that this peritubular response is harmful to the tubule. In contrast, the authors came to the conclusion that the

NU

interstitial inflammation supports the recovery of the tubule.

MA

They argue as follows: In early stages of acute tubular injury, tubules undergo expansion caused by elevated intratubular pressures. Myofibroblasts gather around such tubules in a

TE

D

circular arrangement tightly fixing to the TBM. These contractile cells might counteract further expansion of the tubule and

CE P

thereby favour tubular recovery. Our results in the UUO model [24] and observations in drug-induced tubular injury in rat distal

AC

tubules by thiazides [30,31], by lithium [36] and thick ascending limbs in a transgenic mouse model for uromodulin-associated kidney diseases [35], are in full agreement with these observations

Signalling underlying the tubulointerstitial phenomena To discuss this issue it is necessary to distinguish between signals to the interstitium and those directed to tubular cells. Many studies were undertaken to figure out the factors secreted by injured and/or regenerating tubule cells acting on the 18

ACCEPTED MANUSCRIPT surroundings and being likely responsible for the peritubular reactions. Recent reviews by Venkatachalam and colleagues in

T

2010 [2] and Bonventre and Yang in 2011 [6] have discussed

IP

this issue in detail. A confusingly large array of signalling

SC R

pathways are involved and there is no certain knowledge

NU

whether they are protective, injurious or restorative.

There is agreement that the interstitial process is started by

MA

paracrine stimulation from injured tubular cells. The main factor is the production of TGFß [2,5,6,18,41,45-47]

D

TGFß induced loss of PTEN appears to contribute to the failure

TE

of regenerating epithelial cells to redifferentiate, thereby causing the retention of proliferative signalling, and in this

CE P

fashion, give rise to the production of profibrotic peptides [5]. TGF stimulates fibrosis not only via its direct effects on

AC

fibroblasts but additionally by inducing the production of other molecules like Notch [48], and CTGF [49] Yang 2012) or PDGF [2,43,46]. In addition, TGFß mediates upregulation of osteopontin (OPN) expression in injured epithelial cells [50] and triggers a focal inflammatory process with immigration of neutrophils (Fig. 4a), mononuclear cells [51,52] and macrophages [53]. The resident dendritic cells adopt a proinflammatory phenotype and activate T-cells, but do not directly contribute to fibrosis [54]. These factors taken together initiate the proliferation of interstitial fibroblasts and enhance the 19

ACCEPTED MANUSCRIPT differentiation of fibroblasts into myofibroblasts [55] with the

T

effects described above

IP

Additionally, changes in the capillary network are widely

SC R

believed to be involved. As a consequence of tubular injury, peritubular capillaries may be deprived from survival factors of tubular origin, first of all VEGF [2] leading to capillary

NU

rarefaction. The potentially resulting focal hypoxia is thought to

MA

contribute to inflammation and fibrosis. However, no direct evidence of a vascular contribution is available. Inflammation and the emergence of myofibroblasts in the peritubular

TE

D

interstitium have been observed already after one to two days following initiation of injury in UUO [24] and injection of uranyl

CE P

acetate [32] or a of thiazide diuretic [30,31]. Remodelling of the capillary network would hardly have taken place in such short

AC

periods of time. In addition, after injury with uranyl acetate or thiazide the interstitium was affected precisely around the specifically targeted nephron segment, the S3 segment of the proximal tubule and the distal convoluted tubule, respectively. Any signal produced by peritubular capillaries would have affected a broader area.

Whereas considerable effort has been devoted to identify signals from injured tubules affecting the interstitium, little is known about signals from the interstitium, in particular from

20

ACCEPTED MANUSCRIPT fibroblasts or myofibroblasts, which might influence the fate of

T

injured tubular cells.

IP

Much less is known about signals that regulate the regeneration

SC R

/ degeneration of injured tubular cells. Above we discussed the hypothesis that failed differentiation and growth arrest of a certain fraction of regenerating tubular cells after an acute

NU

tubular injury is responsible for the maintenance of the

MA

tubulointerstitial process [2].. Here the question arises, whether this failure depends exclusively on signalling inside the injured

TE

D

tubule or whether signals from outside may also be relevant.

No candidates for signals from myofibroblasts to the tubule

CE P

have ever been proposed. As noted by Lan and colleagues [5] individual tubule cells showing signs of failed differentiation and

AC

growth arrest can be found amidst regenerating cells . Such a pattern cannot be the consequence of signalling from outside the tubule but appears to be based on inherent properties of those cells. Paracrine and/or autocrine effects of TGF might be involved [5].

Relevance of fibrosis for progression of CKD?

Above we considered fibrosis as a healing process, which results in the formation of a scar in the place of a lost nephron or supports the healing of a nephron. This concept contrasts 21

ACCEPTED MANUSCRIPT with the prevalent view that renal fibrosis has a progressive potential which significantly contributes to the deterioration of

T

the kidney in CKD [56,57]. Here we will discuss whether fibrosis

IP

may acquire self-sustaining properties contributing to the

SC R

destruction of so far healthy or partially damaged nephrons or whether any sustained fibrotic activity is the consequence of persisting disease processes within the nephron. Let us start

MA

NU

with fibrosis subsequent to a primary glomerular disease.

Relevance of fibrosis for the progression of CKD subsequent to glomerular diseases

TE

D

The hypothesis of an active contribution of fibrosis to progression is essentially based on the observation that the

CE P

decline in renal function in chronic renal disease correlates more closely with interstitial fibrosis than with glomerular

AC

damage [58-61].

This correlation is taken as argument in favour of a genuine interstitial mechanism of progression. Thereby one overlooks a crucial fact: As animal models [1] and studies in the aging human kidneys show [62,63] the remnants of a degenerating 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. 22

ACCEPTED MANUSCRIPT More recent studies clearly show that the decline in renal function in chronic renal disease correlates best with the

T

number of remaining nephrons [64-66]. Thus, the evidence for

IP

the fibrosis hypothesis from these studies is far from being

SC R

conclusive.

Our own observations in a variety of animal models do not

NU

support the hypothesis that fibrosis may become a self-

MA

sustained process that injures healthy nephrons. We consistently found that tubule segments of healthy nephrons, even if completely surrounded by degenerating tubules or

TE

D

embedded in fibrotic tissue, preserve their normal structure. Figure 6 shows an area of extensive nephron degeneration

CE P

from an experimental fibrosis model (overexpression of TGFß; [18] that contains a single structurally intact nephron completely

AC

trapped into an area of tubulo-interstitial fibrosis. Surely, such examples do not prove that on the long run such a hostile environment will not exert damaging effects to this nephron. However, in autopsies or biopsies of endstage kidneys groups of structurally intact tubular profiles are frequently encountered surrounded by scar tissue. If harmful signals from fibrotic areas or local hypoxia due to capillary rarefication (as frequently suggested) were the major cause of tubular injury, all nephrons in that area would display a similar degree of damage.

23

ACCEPTED MANUSCRIPT The pathophysiological situation in a stage of advanced fibrosis consisting of proliferating and fibrotic areas, of dead, of

T

damaged and of some intact (but mostly hypertrophied)

IP

nephrons is complex and chaotic. The chance to clearly

SC R

decipher profibrotic and antifibrotic, destructive and protective mechanisms is small, at best, as is the chance for any successful therapeutic interventions. The fight against

NU

progression of glomerular disease has to start much earlier at

MA

stages when the process is still confined to the glomerulus. Regarding glomerular injuries a vicious cycle has been documented, in which podocyte loss increases the probability of

TE

D

further podocyte loss [1]. That mechanism underlies the glomerular overload hypothesis for progression, first formulated

CE P

by Brenner [67,68]. An equally evidence-supported hypothesis is not available for fibrosis as a self-sustained damaging

AC

mechanism.

Relevance of fibrosis in the progression of a tubular injury to CKD The issue of how acute kidney injury may enter a chronic course eventually leading to end-stage renal failure is currently of great interest [69]. As discussed above, the accumulation of myofibroblasts around injured tubules during AKI is not likely to provoke further tubular injury but rather plays a supportive role. Indeed, as may be seen from the work of Fujigaki and colleagues [32] and others (see above) such tubules may 24

ACCEPTED MANUSCRIPT undergo regeneration accompanied by the disappearance of

T

the peritubular changes, i.e. the myofibroblasts.

IP

On the other hand these pathologies may progress to the loss

SC R

of the entire nephron and to CKD as shown in a recent study by Grgic and colleagues [19]. In an inducible transgenic mouse model a single toxin insult to proximal tubules was promptly

NU

followed by peritubular inflammation, tubule cell proliferation

MA

and recovery. However, three insults at one-week intervals led to a progressing tubulo-interstitial disease that encroached upon the glomerulus. Glomerulosclerosis significantly correlated

TE

D

with interstitial fibrosis. The authors suggested several mechanisms responsible for the transfer of the disease on the

CE P

glomerulus, among them that obstructive damage to the initial segment of the proximal tubule initiated sclerosis development

AC

in the glomerulus, leading to sclerotic, likely atubular glomeruli as commonly observed in obstructive nephropathy [21]. In case that this happens in a sufficiently large fraction of nephrons overload will be imposed on the remaining nephrons starting the vicious cycle to progression to CKD proposed by Brenner [67,68]. In that way, an originally tubular disease may enter a progressing course due to the involvement of glomeruli.

This discussion leads to the insight that the regenerative potential is fundamentally different between foci of "tubulointerstitial fibrosis" derived from a primary glomerular 25

ACCEPTED MANUSCRIPT disease and those of primary tubular origin. Descriptively, both cases of persistent "tubulointerstitial fibrosis" are similar.

T

However, they differ with respect to the underlying mechanisms

IP

and to the possible outcome. Primary tubular injuries leave a

SC R

chance of healing and the transient fibrosis seems to play a supportive role. Tubular injuries derived from a glomerular disease do not have this chance: they fully depend on the

NU

disease in the corresponding glomerulus. This is not expected

MA

to revert since full structural and functional recovery of a sclerotic or crescentic injury is unlikely to occur. On the contrary, such glomerular lesions tend to progress. Since

TE

D

progression proceeds at different rates in different glomeruli some nephrons will still be operative while others have

CE P

degenerated. The maintenance of a residual function obviously requires the structural integrity of the still perfused tubules.

AC

Fibrosis may play an important role in that respect. Nephrons are fragile microanatomical entities, which are not supported by connective tissue boundaries as found in most other organs (e. g. glands, lung). The tubular coils of adjacent nephrons come very close to each other and to glomeruli and the shape of a nephron is largely maintained by its relationships to neighbouring nephrons. The tenuous network of interstitial fibroblasts acts as a link between neighbouring structures. How to maintain a mechanical support for a nephron when adjacent nephrons undergo decomposition? It seems plausible that fibrosis provides a stable three-dimensional framework 26

ACCEPTED MANUSCRIPT permitting the survival and residual function of still intact or

IP

T

partially damaged nephrons.

SC R

Concluding remarks

NU

Renal fibrosis is consistently associated with chronic renal disease. It develops in conjunction with tubular degeneration,

MA

therefore the generally used but vague term "tubulointerstitial fibrosis”. This designation obscures the individual roles of the

D

tubular and the interstitial component in this process and led to

TE

the widely accepted hypothesis that fibrosis causes the degeneration of the tubule. However, any reliable evidence

CE P

speaks in favour of the opposite mechanism: the progressing

AC

glomerular and tubular injury drives the fibrosis.

Progression of fibrosis accompanied with a glomerular disease crucially depends on the progression of the injury in the glomerulus. If this process has passed the point of no return, i.e the loss of separation between the tuft and Bowman`s capsule, the only option to control the course of the disease consist of slowing down the progression of the glomerular injury and, consequently, that of the tubule. In case of progression to nephron loss, the resulting replacement of the nephron by fibrosis may well play a protective role for the remaining 27

ACCEPTED MANUSCRIPT nephrons providing them mechanically secured spaces to

T

survive.

IP

The situation of fibrosis subsequent to a primary tubular disease

SC R

is different. The dominating process associated with a tubular injury is regeneration. This is accompanied by a transient peritubular inflammation and fibrosis that appear to support the

NU

recovery of the tubule. Peritubular fibrosis persists when tubules

MA

fail to undergo full regeneration and differentiation. This failure, the causes of which are largely unknown, may eventually lead to the involvement of the glomerulus and to the glomerulus

TE

D

dependent vicious circle resulting in CKD. If a sufficient number of nephrons is concerned, the glomerular damage-dependant

CE P

vicious cycle to progression to CKD will be started. Thus, the entry of a tubular disease into a progressing course likely

AC

depends on its mutation into a glomerular disease.

Acknowledgements The authors would like to thank Brunhilde Hähnel and Hiltraud Hosser for tireless technical and organizational help. The authors gratefully acknowledge the continuous support of our work by the "Gotthard Schettler Gesellschaft fuer Herz- und Kreislauf-forschung" and the "Professor Dr. Karl und Gerhard Schiller Stiftung".

Figure legends 28

ACCEPTED MANUSCRIPT

IP

T

Fig.1 Encroachment of a glomerular injury onto the tubulointerstitium Two injured glomeruli and the corresponding tubules are highlighted in yellow. The crescentic glomerular injury

SC R

(asterisks) encroaches inside the basement membrane (arrows) on the proximal tubule. The tubular epithelium is severely

NU

damaged, most advanced in proximal tubules immediately after the glomerulo-tubular junction (star), less advanced in distal

MA

tubule segments (triangles). Around injured glomeruli and tubules a vivid proliferation of interstitial cells is seen. Tubular

D

profiles belonging to glomeruli with a patent urinary orifice are

TE

intact (circles; verified by tracing in serial sections).

CE P

Thy-1 mediated glomerulonephritis in rat, day 9; 1µm Epon section; bar: 20 µm. Reproduced with permission from Kriz et

AC

al. (2003) [8]

Fig.2 Tubule degeneration A: Overview of a group of degenerating proximal tubules. The severity of epithelial changes increases downstream; upstream segments (1, 2, 3, 4) exhibit epithelial desintegration with cells filled with autophagic vacuoles (stars), in downstream segments (5, 6) the epithelial integrity is still preserved. The surrounding interstitium is expanded, rich in cells and fluid, and contains ample capillaries. Proximal tubules (asterisks) of unaffected nephrons exhibit an intact epithelium. 29

ACCEPTED MANUSCRIPT B: Collapsed profile of proximal tubule in an advanced stage of destruction. Cells contain assemblies of autophagic vacuoles

T

(asterisks) and start to disintegrate (thick arrows).The tubular

IP

basement membrane (shown in yellow) is wrinkled

SC R

(arrowheads). The interstitium contains many myofibroblasts, processes of which (arrows) form an encircling envelope. A: Thy-1 mediated glomerulonephritis in rat, 2 weeks;

NU

transmission electron micrograph (TEM), bar: 3 µm.

MA

Reproduced with permission from Kriz et al. (2003) [8] B: Anti-GBM glomerulonephritis, day 10; bar: 5 µm.

D

Unpublished TEM from [9].

TE

Fig.3 Renal cortical interstitium

CE P

A: Schematic illustration. Interstitial fibroblasts are shown in green, a dendritic cell in yellow, tubules in violet, a capillary is

AC

labelled with C.

Fibroblasts have multiple projections that bridge the interstitial space terminating in focal adhesions at tubules, capillaries and processes of other fibroblasts (arrowheads). They contain thin assemblies of f-actin filaments that run through the cell bodies (located immediately under the plasmalemm) and extend into the processes. Thus, renal fibroblasts form a network that interconnects all structures of renal cortex. Dendritic cells show more rounded cell bodies and lack the dense f-actin

30

ACCEPTED MANUSCRIPT assemblies; their generally thin cell projections are intermingled with fibroblast processes.

T

B: Three-dimensional view of the renal cortical interstitium of rat

IP

by scanning electron microscopy, after digestion of tubular and

SC R

vascular basement membranes. Fibroblast cell bodies (S) give rise to attenuated and perforated leaflet-like processes that touch each other at many sites. Dendritic cells (asterisk),

NU

enclosed into the network of fibroblasts, form broad cytoplasmic

MA

projections (arrows) as well as filiform processes (arrow heads).

PT, proximal tubule; C, capillary; bar: 10 µm.

TE

D

Reproduced with permission from Takahashi-Iwanaga (1991)

Fig.4

CE P

[70].

AC

Rat renal cortex after 4 days of ureter ligation. Immunofluorescence: ecto-5- nucleotidase (5-NT)— red , smooth muscle actin (SMA)—green; chromatin staining by DAPI—blue channel. 5’NT labels all cortical interstitial fibroblasts, the brush border of proximal tubules (P), and intercalated cells in a connecting tubule (CN). In areas around proximal tubules with advanced degeneration (asterisks) SMA is strongly upregulated in fibroblasts while 5’NT is faintly detectable; in fibroblasts in areas with less advanced injury of proximal tubules (P) SMA is 31

ACCEPTED MANUSCRIPT weakly upregulated while 5’NT is distinctly detcetable: A, glomerular arteriole; 3 µm cryostat section; bar: 50 µm.

T

Reproduced with permission from Picard et al. (2008) [24]

SC R

Peritubular response to tubular injury

IP

Fig. 5

Tubular damge and interstitial response induced by thiazide diuretics; thiazides inhibit specifically the neutral sodium co-

NU

transporter (NCC) in the apical plasma membrane of cells of the

MA

distal convoluted tubule (DCT). In rats the treatment induces damage exclusively in DCT profiles [30] A: DCT profile (D) 24 hours after a single dose of thiazide-like

TE

D

diuretic metolazon®; tubular injury becomes evident by upregulation of osteopontin (red; thin arrrows) in DCT cells. At

CE P

this stage of injury neutrophil leucocytes (labelled by OX 1 antibody directed against common leucocyte antigen; green;

AC

thick arrow) adhere to the injured cells. 1 µm cryostat section; bar: 10 µm.

B: After 36 hours cells in the DCT (D) are apoptotic and the peritubular interstitial fibroblasts have transformed to myofibroblasts (arrows) and encapsulate the injured tubule. 1 µm Epon section; bar: 10 µm. C: After 72 hours massive up-regulation of SMA in the interstitium immediately around the injured tubule (D); the strength of fluorescence for SMA decreases towards more distant areas (arrows). 3 µm cryostat section; bar: 20 µm.

32

ACCEPTED MANUSCRIPT A and B: unpublished micrographs from [30] ; C: reproduced

T

with permission from Kaissling and Le Hir (2008) [27]

IP

Fig. 6

SC R

Advanced Fibrosis

A: Area of extensive nephron degeneration and fibrosis housing a single intact nephron (shown in yellow). The higher

NU

magnification in (B) shows that the glomerulus, the glomerulo-

MA

tubular junction and the corresponding tubule segments do not exhibit any damage.

Over-expression of TGFβ in renal tubules in mouse [18] , after 6

TE

D

weeks. 1 µm Epon section; bars: A: 40 µm, B: 20 µm.

CE P

Unpublished light micrographs from [18]

Reference List

AC

[1] W. Kriz and M. Le Hir, Pathways to nephron loss starting from glomerular diseases - Insights from animal models, Kidney Int, 67 (2005) 404-419. [2] M. Venkatachalam, K. Griffin, R. Lan, H. Geng, P. Saikumar, and A. Bidani, Acute kidney injury: a springboard for progression in chronic kidney disease, Am J Physiol Renal Physiol, 298 (2010) F1078-F1094. [3] K. Nath, Tubulointerstitial changes as a major determinant in the progression of renal damage, Am J Kidney Dis, 20 (1992) 1-17. [4] G. Hill, Hypertensive nephrosclerosis, Curr Opin Nephrol Hypertens, 17 (2008) 266-270. [5] R. Lan, H. Geng, A. Polichnowski, P. Singha, P. Saikumar, D. McEwen, K. Griffin, R. Koesters, J. Weinberg, A. Bidani, W. Kriz, and K. Venkatachalam, PTEN loss defines a TGF-ß induced tubule phenotype of failed differentiation and JNK signaling during 33

ACCEPTED MANUSCRIPT renal fibrosis, Am J Physiol Renal Physiol, 302 (2012) F1210F1223.

T

[6] J. Bonventre and L. Yang, Cellular pathophysiology of ischemic acute kidney injury, J Clin Invest, 121 (2011) 4120-4221.

SC R

IP

[7] M. Kretzler, I. Koeppen-Hagemann, and W. Kriz, Podocyte damage is a critical step in the development of glomerulosclerosis in the uninephrectomized desoxycorticosterone rat., Virchows Arch A Pathol Anat Histopathol, 425 (1994) 181-193.

NU

[8] W. Kriz, B. Hähnel, H. Hosser, T. Ostendorf, B. Kränzlin, N. Gretz, F. Shimizu, and J. Floege, Pathways to recovery and loss of nephrons in anti-Thy-1 nephritis, J Am Soc Nephrol, 14 (2003) 1904-1926.

D

MA

[9] M. Le Hir, C. Keller, V. Eschmann, B. Hähnel, H. Hosser, and W. Kriz, Podocyte bridges between the tuft and Bowman's capsule: An early event in experimental crescentic glomerulonephritis, J Am Soc Nephrol, 12 (2001) 2060-2071.

CE P

TE

[10] G. Remuzzi, A unifying hypothesis for renal scarring linking protein trafficking to the different mediators of injury, Nephrol Dial Transplant, 15 (2000) 58-60.

AC

[11] J. Guo, A. Marlier, H. Shi, A. Shan, T. Ardito, Z. Du, M. Kashgarian, D. Krause, D. Biemesderfer, and L. Cantley, Increased tubular proliferation as an adaptive response to glomerular albuminuria, J Am Soc Nephrol, 23 (2012) 429-437. [12] F. Theilig, W. Kriz, T. Jerichow, P. Schrade, B. Hähnel, T. Willnow, M. Le Hir, and S. Bachmann, Abrogation of protein uptake through Megalin-deficient proximal tubules does not safeguard against tubulointerstitial injury, J Am Soc Nephrol, 18 (2007) 1824-1834. [13] L. Fine and J. Norman, Chronic hypoxia as a mechanism of progression of chronic kidney diseases: from hypothesis to novel therapeutics, Kidney Int, 74 (2008) 867-872. [14] A. Bohle, S. Mackensen-Haen, and M. Wehrmann, Significance of post-glomerular capillaries in the pathogenesis of chronic renal failure, Kidney Blood Press Res, 19 (1996) 191-195.

34

ACCEPTED MANUSCRIPT [15] H. Rollhäuser, W. Kriz, and W. Heinke, Das Gefässsystem der Rattenniere, Z Zellforsch Mikrosk Anat, 64 (1964) 381-403.

IP

T

[16] F. Theilig, A. Enke, B. Scolari, D. Polzin, S. Bachmann, and R. Koesters, Tubular Deficiency of von Hippel-Lindau Attenuates Renal Disease Progression in Anti-GBM Glomerulonephritis, Am J Pathol, 179 (2011) 2177-2188.

SC R

[17] L. Li, D. Zepeda-Orozco, R. Black, and F. Lin, Autophagy is a component of epithelial cell fate in obstructive uropathy, Am J Pathol, 176 (2010) 1767-1778.

MA

NU

[18] R. Koesters, B. Kaissling, M. Le Hir, N. Picard, F. Theilig, R. Gebhardt, A. Glick, B. Hähnel, H. Hosser, H. Gröne, and W. Kriz, Tubular overexpression of transforing growth factor-ß1 induces autophagy and fibrosis but not mesenchymal transition of renal epithelial cells, Am J Pathol, 177 (2010) 632-643.

TE

D

[19] I. Grgic, G. Campanholle, V. Bijol, C. Wang, V. Sabbisetti, T. Ichimura, B. Humphreys, and J. Bonventre, Targeted proximal tubule injury triggers interstitial fibrosis and glomerulosclerosis, Kidney Int, 82 (2012) 172-183.

CE P

[20] N. Marcussen, Atubular glomeruli in chronic renal disease, Curr Top Pathol, 88 (1995) 145-174.

AC

[21] R. Chevalier and M. Forbes, Generation and evaluation of atubular glomeruli in the progression of renal disorders, J Am Soc Nephrol, 19 (2008) 197-206. [22] H. Gröne, E. Warnecke, and C. Olbricht, Characteristics of renal tubular atrophy in experimental renovascular hypertension: a model of kidney hibernation, Nephron, 72 (1996) 243-252. [23] W. Kriz, B. Kaissling, and M. Le Hir, Epithelial-mesenchymal transition (EMT) in kidney fibrosis: fact or fantasy?, J Clin Invest, 121 (2011) 468-474. [24] N. Picard, O. Baum, A. Vogetseder, B. Kaissling, and M. Le Hir, Origin of renal myofibroblasts in the model of unilateral ureter obstruction in the rat, Histochem Cell Biol, 130 (2008) 141-155. [25] B. Humphreys, Lin SL, A. Kobayashi, T. Hudosn, B. Nowlin, J. Bonventre, M. Valerius, A. McMahon, and J. Duffield, Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis, Am J Pathol, 176 (2010) 85-97. 35

ACCEPTED MANUSCRIPT [26] J. Duffield, K. Park, L. Hsiao, V. Kelley, D. Scadden, T. Ichimura, and J. Bonventre, Restoration of tubular epithelial cells during repair of the postischemic kidney occurs independently of bone marrow-derived stem cells, J Clin Invest, 115 (2005) 1743-1755.

SC R

IP

T

[27] B. Kaissling and M. Le Hir, The renal cortical interstitium: morphological and functional aspects, Histochem Cell Biol, 130 (2008) 247-262. [28] J. Tomasek, G. Gabbiani, B. Hinz, C. Chaponnier, and R. Brown, Myofibroblasts and mechano-regulation of connective tissue remodelling, Nat Rev Mol Cell Biol, 3 (2002) 349-363.

MA

NU

[29] K. Chugh, B. Sharma, P. Singhal, K. Das, and Datta BN, Acute renal failure following copper sulphate intoxication, Postgrad Med J, 53 (2012) 18-23.

D

[30] J. Loffing, D. Loffing-Cueni, I. Hegyi, M. Kaplan, S. Hebert, M. Le Hir, and B. Kaissling, Thiazide treatment of rats provokes apoptosis in distal tubule cells, Kidney Int, 50 (1996) 1180-1190.

CE P

TE

[31] M. Le Hir, I. Hegyi, D. Loffing-Cueni, J. Loffing, and B. Kaissling, Characterization of renal interstitial fibroblast-specific protein 1/S100A4-positive cells in healthy and inflamed rodent kidneys., Histochem Cell Biol, 123 (2005) 335-346.

AC

[32] Y. Fujigaki, Y. Muranaka, D. Sun, T. Goto, H. Zhou, M. Sakakima, H. Fukasawa, K. Yonemura, T. Yamamoto, and A. Hishida, Transient myofibroblast differentiation of interstitial fibroblastic cells relevant to tubular dilatation in uranyl acetate-induced acute renal failure in rats, Virchows Arch, 446 (2005) 164-176. [33] R. Miller, R. Tadagavadi, G. Ramesh, and W. Reeves, Mechanisms of Cisplatin nephrotoxicity, Toxins, 2 (2010) 24902518. [34] H. Gröne, K. Weber, E. Gröne, U. Helmchen, and M. Osborne, Coexpression of Keratin and Vimentin in damaged and regenerating tubular epithelia of the kidney, Am J Pathol, 129 (1987) 1-8. [35] S. Bernascone, J. Erriquez, M. Ferraro, A. Genazzani, and C. Distasi, Novel adenosine and cAMP signalling pathways in migrating glial cells, Cell Calcium, 48 (2010) 83-90.

36

ACCEPTED MANUSCRIPT [36] M. Zhu, M. Broe, and E. Nouwen, Vimentin expression and distal tubular damage in the rat kidney, Exp Nephrol, 4 (1996) 172-183.

IP

T

[37] A. Vogetseder, T. Palan, D. Bacic, B. Kaissling, and M. Le Hir, Proximal tubular epithelial cells are generated by division of differentiated cells in the healthy kidney, Am J Physiol Cell Physiol, 292 (2007) 807-813.

SC R

[38] A. Vogetseder, N. Picard, A. Gaspert, M. Walch, B. Kaissling, and M. Le Hir, Proliferation capacity of the renal proximal tubule involves the bulk of differentiated epithelial cells, Am J Physiol Cell Physiol, 294 (2008) 22-28.

MA

NU

[39] M. Langworthy, B. Zhou, M. de Caestecker, G. Moeckel, and H. Baldwin, NFATc1 identifies a population of proximal tubule cell progenitors, J Am Soc Nephrol, 20 (2009) 311-321.

TE

D

[40] B. Smeets, P. Boor, H. Dijkman, S. Sharma, P. Jirak, F. Mooren, K. Berger, J. Bornemann, I. Gelman, J. Floege, J. van der Vlag, J. Wetzels, and M. Moeller, Proximal tubular cells contain a phenotypically distinct, scattered cell population involved in tubular regeneration, J Pathol, nn (2012) nn.

CE P

[41] L. Yang, T. Besschetnova, C. Brooks, J. Shah, and J. Bonventre, Epithelial cell cycle arrest in G2/M mediates kidney fibrosis after injury, Nat Med, 16 (2010) 535-543.

AC

[42] D. Sun, Y. Fujigaki, T. Fujimoto, K. Yonemura, and A. Hishida, Possible involvement of myofibroblasts in cellular recovery of uranyl acetate-induced acute renal failure in rats, Am J Pathol, 157 (2000) 1321-1335. [43] T. Suzuki, M. Kimura, M. Asano, Y. Fujigaki, and A. Hishida, Role of atrophic tubules in development of interstitial fibrosis in microembolism-induced renal failure in rat, Am J Pathol, 158 (2001) 75-85. [44] D. Sun, Y. Fujigaki, T. Fujimoto, T. Goto, K. Yonemura, and A. Hishida, Mycophenolate Mofetil inhibits regenerative repair in uranyl acetate-induced acute renal failure by reduced interstitial cellular response, Am J Pathol, 161 (2002) 217-227. [45] H. Lin, Transforming growth factor-ß and the kidney revisited: Introduction, Semin Nephrol, 32 (2012) 225-227.

37

ACCEPTED MANUSCRIPT

IP

T

[46] H. Geng, R. Lan, P. Singha, A. Gilchrist, P. Weinreb, S. Violette, J. Weinberg, P. Saikumar, and K. Venkatachalam, Lysophosphatidic acid increases proximal tubule cell secretion of profibrotic cytokines PDGF-B and CTGF through LPA2- and Gqmediated Rho and vß6 integrin-dependent activation of TGF-ß, Am J Pathol, 181 (2012) 1236-1249.

SC R

[47] K. Spurgeon, D. Donohoe, and D. Basile, Transforming growth factor-beta in acute renal failure: receptor expression, effects on proliferation, cellularity, and vascularization after recovery from injury, Am J Physiol Renal Physiol, 288 (2005) F568-F577.

MA

NU

[48] B. Bielesz, Y. Sirin, H. Si, T. Niranjan, A. Gruenwald, S. Ahn, H. Kato, J. Pullman, M. Gessler, V. Haase, and K. Susztak, Epithelial Notch signaling regulates interstitial fibrosis development in the kidneys of mice and humans, J Clin Invest, 120 (2010) 4040-4054.

TE

D

[49] H. Okada, T. Kikuta, T. Kobayashi, T. Inoue, Y. Kanno, M. Takigawa, T. Sugaya, J. Kopp, and H. Suzuki, Connective tissue growth factor expressed in tubular epithelium plays a pivotal role in renal fibrosis, J Am Soc Nephrol, 16 (2005) 133-143.

CE P

[50] M. Mazzali, T. Kipari, V. Ophascharoensuk, J. Wesson, R. Johnson, and J. Hughes, Osteopontin--a molecule for all seasons, QJM, 95 (2002) 3-13.

AC

[51] G. Kinsey, L. Li, and M. Okusa, Inflammation in acute kidney injury, Nephron Exp Nephrol, 109 (2008) 102-107. [52] A. Chung and H. Lan, Chemokines in renal injury, J Am Soc Nephrol, 22 (2011) 802-809. [53] K. Rock, E. Latz, F. Ontiveros, and H. Kono, The sterile inflammatory response, Annu Rev Immunol, 28 (2010) 321-342. [54] S. Snelgrove, J. Kausman, C. Lo, C. Lo, P. Coates, M. Hickey, S. Holdsworth, C. Kurts, D. Engel, and A. Kitching, Renal dendritic cells adopt a pro-inflammatory phenotype in obstructive uropathy to activate T cells but do not directly contribute to fibrosis, Am J Pathol, 180 (2012) 91-103. [55] A. Desmouliere, C. Chaponnier, and G. Gabbiani, Tissue repair, contraction, and the myofibroblast, Wound Repair Regen, 13 (2005) 7-12. 38

ACCEPTED MANUSCRIPT [56] W. Bechtel, S. McGoohan, E. Zeisberg, G. Müller, H. Kalbacher, D. Salant, C. Müller, R. Kalluri, and M. Zeisberg, Methylation determines fibroblast activation and fibrogenesis in the kidney, Nat Med, 16 (2010) 544-551.

IP

T

[57] Y. Liu, Renal fibrosis: New insights into the pathogenesis and therapeutics, Kidney Int, 69 (2012) 213-217.

SC R

[58] R. Ridson, J. Sloper, and H. de Wardener, Relationship between renal function and histological changes found in renal-biopsy specimens from patients with persistent glomerular nephritis, Lancet, 2 (1968) 363-366.

MA

NU

[59] A. Bohle, D. Glomb, K. Grund, and S. Mackensen, 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 (1977) 221-232.

TE

D

[60] G. d'Amico, F. Ferrario, and M. Rastaldi, Tubulointerstitial damage in glomerular diseases: Its role in the progression of renal damage, Am J Kidney Dis, 26 (1995) 124-132.

CE P

[61] J. Cameron, Tubular and interstitial factors in the progression of glomerulonephritis, Pediatr Nephrol, 6 (1992) 292-303.

AC

[62] Y. Kanwar, Biology of disease. Biophysiology of glomerular filtration and proteinuria, Lab Invest, 51 (1984) 7-21. [63] G. Yang and A. Morrison, Three large dissectable rat glomerular models reconstructed from wide-field electron micrographs, Anat Rec, 196 (1980) 431-440. [64] I. Bajema, E. Hagen, J. Hermans, L. Noel, R. Waldherr, F. Ferrario, F. van der Woude, and J. Bruijn, Kidney biopsy as a predictor for renal outcome in ANCA-associated necrotizing glomerulonephritis, Kidney Int, 56 (1999) 1751-1758. [65] M. Lubran, Renal function in the elderly, Ann Clin Lab Sci, 25 (1995) 122-133. [66] T. Bendtsen and J. Nyengaard, The number of glomeruli in type 1 (insulin-dependent) and type 2 (non-insulin-dependent) diabetic patients, Diabetologia, 35 (1992) 844-850.

39

ACCEPTED MANUSCRIPT [67] T. Hostetter, J. Olson, H. Rennke, M. Venkatachalam, and B. Brenner, Hyperfiltration in remnant nephrons:a potentially adverse response to renal ablation, Am J Physiol, 241 (1981) F85-F93.

IP

T

[68] B. Brenner, E. Lawler, and H. Mackenzie, The hyperfiltration theory: a paradigma shift in nephrology, Kidney Int, 49 (1996) 1774-1777.

SC R

[69] K. Leung, M. Tonelli, and M. James, Chronic kidney disease following acute kidney injury - risk and outcomes, Nat Rev Nephrol, 9 (2013) 77-85.

AC

CE P

TE

D

MA

NU

[70] H. Takahashi-Iwanaga, The three-dimensional spaces of the kidney, Cell Tissue Res, 264 (1991) 269-281.

40

Figure 1

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

41

Figure 2A

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

42

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 2B

43

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 3A

44

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 3B

45

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 4

46

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 5A

47

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 5B

48

Figure 5C

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

49

Figure 6A

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

50

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

Figure 6B

51

ACCEPTED MANUSCRIPT Highlights

Fibrosis associated with a tubular injury is supportive for recovery

IP

T

Fibrosis associated with a glomerular disease fully depends on the progression of the glomerular injury.

SC R

Fibrosis provides secured spaces for intact or partially injured nephrons to survive

AC

CE P

TE

D

MA

NU

Fibrosis is no promoter of progression, in contrast, fibrosis is protective

52