Main

The pathology of kidney diseases is exceptionally diverse because of the complex cellular composition and anatomy of the nephrons. It is largely characterized by a mix of intrinsic injurious pathomechanisms destroying some nephrons together with concomitant healing responses that try to stabilize the other nephrons. Still, a variety of different triggers can induce similar histopathological lesions in the kidney, for example, glomerular crescents, focal segmental glomerulosclerosis (FSGS), mesangial proliferation, thrombotic microangiopathy, tubular necrosis, or interstitial fibrosis. Therefore, a kidney biopsy often displays non-specific lesions, which require clinical and laboratory correlation for a definite diagnosis. In addition, kidney pathology manifests with distinct patterns of structural abnormalities, raising the question of potential causative hierarchical relationships. In this review, we address this topic taking the approach of evolutionary medicine to better understand why kidney injury generates distinct histological lesions and how they are causatively related. This conceptual approach appears superior in delineating the most likely molecular and cellular targets to therapeutically modulate kidney disease for better outcomes. In particular, we focus on the molecular links between distinct danger response programs as these may be most powerful in modulating complex disease processes.

DANGER RESPONSE PROGRAMS SAVE LIFE, USUALLY

Both plants and animals control traumatic injuries either by regeneration or repair.1 Traumatic injuries disrupt epithelial barriers, which imply a number of dangers to homeostasis, including the risks of fluid loss and microbial infection. These dangers required control mechanisms (constraints) from early on in the evolution of multicellular organisms and those that were maintained during evolution up to humans obviously resulted in significant survival benefits that outweigh collateral injuries or tissue remodeling. Such trade-offs under certain conditions can translate into pathomechanisms and drive disease progression. In the following paragraphs, we will introduce the four major danger response programs: coagulation, inflammation, re-epithelialization, and mesenchymal repair and discuss their pathogenic roles in kidney disease.2

Vascular Injury Induces Coagulation Rapidly to Minimize Bleeding

Traumatic injury of blood vessels causes bleeding. Clotting minimizes the risk of hemorrhagic shock, eg, hemolymph aggregation in arthropods, and presents as a more complex interplay of activated endothelial cells, coagulation factors, and platelets in vertebrates.3 Although clotting minimizes short-term risks, intravascular coagulation or thromboembolism can create subsequent side effects (conflicts), such as tissue hypoperfusion and ischemia.4, 5

Epithelial Barrier Injuries and Coagulation Induce Inflammation as a Functional Barrier Against infection

The evolution of multicellular organisms is strongly influenced by the balance between microbial virulence factors and antimicrobial host defense.6 Wounds disrupt barriers that protect from pathogen entry; vice versa, wounding implies the risk for infection. Hence, inflammatory response intends to rapidly install a functional barrier against pathogen entry and spreading.7, 8 Pathogen-derived pathogen-associated molecular patterns (PAMPs) and intrinsic cell death-related damage-associated molecular patterns (DAMPs) activate the identical pattern-recognition receptors either in infectious or in sterile inflammation.9, 10, 11, 12, 13 There is a close link between clotting (the process that changes blood from liquid to gel) and inflammation, as platelet activation and aggregation induces release of cytokines and chemokines from platelet granules that promote recruitment of leukocytes and local inflammation, a process referred to as immunothrombosis.14, 15, 16, 17 For example, platelet-derived chemokines foster neutrophil recruitment that eventually undergo NETosis, a form of pathogen-induced cell death during which neutrophil extracellular traps (NETs) are released and which further fuels into inflammation.18, 19, 20 The inflammatory response triggers immunopathology and loss of parenchymal cells similar to pyoderma gangrenosum but the same occurs inside the kidney.21 In sepsis, massive activation of cytokine release (cytokine storm) is the major factor and cause of mortality in early sepsis, while the compensatory downregulation of innate immunity accounts for secondary infection-related mortality in late sepsis.22

Epithelial Injury Requires Re-Epithelialization to Minimize Fatal Infection And Fluid Loss

Epithelial injuries require rapid regeneration of the barrier to limit fluid losses and pathogen entry similar to skin burns or ulcerative enteritis.23, 24 Similarly, glomerular epithelial defects cause proteinuria. Furthermore, epithelial transport is another important function of the epithelium to facilitate nutrient absorption (gut), gas exchange (lungs), or solute secretion/reabsorption (kidney). Upon injury, rapid restoration is needed to maintain the function of the corresponding organ, eg, by signals inducing re-epithelialization from wound edges. Already coagulation produces mitogenic factors activating proliferation of surviving epithelial cells.25 Mediators that link coagulation, inflammation, and re-epithelialization include growth factors (epidermal growth factor (EGF), FGFs, hepatocyte growth factor (HGF), TGF-β), cytokines (IL-6, IL-17, IL-22), chemokines (fractalkine, CXCL10) and microRNAs triggering proliferation of epithelial cells with regenerative capacities.26, 27, 28, 29, 30, 31, 32 In the past years, an important contribution of miRNA to cutaneous wound healing has been described.33 These miRNAs are involved in different physiological processes, such as human monocyte/macrophage differentiation (miR-424), Toll-like receptor (TLR)-associated signaling events (miR-147), collagen deposition, and TGF-β-mediated wound contraction (miR-21) or modulating p63 expression (miR-203) during keratinocyte differentiation and in epithelial development.34, 35, 36, 37 Modulating the expression of these miRNA will allow the development of new therapeutic opportunities in wound healing.

The intrinsic regenerative capacity for re-epithelialization largely relates to local progenitor cells committed to the specific epithelial lineage phenotype.23, 38, 39 Insufficient re-epithelialization is associated with persistence of chronic wounds, increasing the risk for infection. In contrast, uncoordinated epithelial hyperplasia can become a pathological mechanism, such as in glomerular crescent formation.40

Tissue Defects Induce Mesenchymal Repair to Restore Tissue Integrity

The reconstitution of parenchymal cell losses requires stabilizing scaffolds that guide parenchymal reconstitution to rebuild structural organization. This scaffold is created by mesenchymal elements within the tissue, eg, fasciae, periosteum, nerve sheaths, vascular pericytes, or interstitial fibroblasts. Therefore, epithelial growth factors driving epithelial healing are released together with mesenchymal growth factors that drive a concurrent mesenchymal healing response that may be transient, depending on epithelial healing completion. If epithelial healing remains incomplete, persistent release of mesenchymal growth factors can promote fibrosis.41 For example, insufficient reconstitution of renal epithelial cells is associated with mesenchymal healing. Epithelial–mesenchymal transition (EMT) and arrest in the G2/M phase of the cell cycle is associated with pro-fibrotic TGF-β cytokine secretion.42 Collagen-producing fibroblasts accumulate mainly via recruitment from bone marrow-derived precursors or by proliferation of resident fibroblasts transforming to myofibroblasts,43, 44, 45 while the mesenchymal transition of pericytes or endothelial cells generate a minor contribution to the collagen-producing cells after injury.46, 47, 48 Hence, interstitial fibrosis supports survival of the renal parenchyma and healing process but also stiffens the tissue, which is called 'sclerosis'.49 Insufficient mesenchymal healing destabilizes tissues during reconstitution, whereas overshooting mesenchymal healing produces lesions, such as keloid in the skin.

WHY DANGER CONTROL PROGRAMS CAN TURN INTO PATHOMECHANISMS

The genetic heterogeneity underlying the spectrum of clinical phenotypes usually follows a bell-shaped pattern. Accordingly, within the population certain individuals will respond to injury either with insufficient or exaggerated coagulation, inflammation, re-epithelialization, and scarring, respectively (Figure 1). For example, genetic defects in complement inhibitors lead to exaggerated complement-related renal inflammation driving the renal pathology of atypical hemolytic uremic syndrome (aHUS).50, 51 Although being a genetic disorder, aHUS does not develop without a trigger of an aHUS episode, usually an infection. Physiological complement activation is necessary for pathogen clearance in most people, but in individuals with a complement-inhibitor deficiency the exaggerated complement activation turns the life-saving immune response into a destructive response (mechanism) damaging the kidney. It is likely that most kidney diseases result from various degrees of imbalance in danger control programs (adaptations with multiple effects), whereas, in healthy conditions, these processes operate in a proper manner during the tissue homeostasis. Thus, in patients with kidney pathology, it is necessary to know why these specific programs are altered.

Figure 1
figure 1

Danger response programs as pathomechanisms of kidney disease. Evolution has developed four adaptive (danger) response programs to survive (traumatic) injuries and infections, ie, clotting, inflammation, and epithelial and mesenchymal healing. In healthy conditions, these operate in an invisible manner during the processes of tissue homeostasis, including epithelial cell turnover or control of the body’s microbiota. Injuries such as a skin laceration activate all four programs to quickly regain homeostasis. However, background genetic variants and environmental influences as well as repetitive or persistent triggers of injury can modulate danger responses and as a by-product cause kidney disease. Patients with kidney pathology experience the effects of one or more imbalances in the danger programs. Examples are given for those kidney diseases in which either insufficient or exaggerated danger responses predominate. It is important to understand the reasons why the specific program is dysregulated. Furthermore, before trying to modulate it, it is important to clarify if a certain dysregulated program is upstream or downstream of nephron loss. For example, if renal scarring is secondary to insufficient epithelial healing, it may be more important to improve epithelial healing rather than blocking fibrosis to protect avoid further nephron loss.

Full size image

Monogenetic disorders may represent the far end of the spectrum, where one genetic defect activates multiple adaptation responses. This implies that in the future our standard renal pathology evaluation should be supplemented with genetic analysis and perhaps algorithms of risk profiles derived from sequencing results for personalized treatment strategies.

The Two Sides of the Coin—Coagulation

Exaggerated clotting as a predominant pathomechanism

Overshooting coagulation is the central pathogenic mechanism of thrombotic microangiopathies. Triggered by an abnormal insult to microvascular endothelial cells, activation of platelets and released plasma coagulation factors can lead to diffuse coagulation in the renal microvasculature.52 As a result, thrombotic microangiopathies cause renal ischemia and necrosis. In contrast, in crescentic glomerulonephritis, glomerular basement membrane (GBM) rupture, leads to plasma leakage, vascular necrosis, and hematuria.53 Such glomerular vascular injuries activate clotting inside glomerular capillaries, which becomes evident by fibrin deposition in capillary loops.54 The link between clotting and inflammation is also named immunothrombosis.16, 55 In addition, in Alport nephropathy progressive GBM disintegration is associated with hematuria, fibrinogen conversion to fibrin, and plasma leakage into Bowman's space.40 During this process, activated platelets release pro-inflammatory and mitogenic factors with numerous effects on glomerular pathology.

Insufficient clotting in kidney disease

Persistent bleeding indicates an insufficient vascular healing by coagulation. Many glomerular diseases are associated with persistent microhematuria, ie, persistent bleeding. Capillary wall injury in few glomeruli can produce hematuria.56, 57 It remains uncertain if persistent hematuria always originates from vascular lesions of the same glomeruli that do not heal or whether clots stop bleeding in some glomeruli, whereas others start to bleed with persistent hematuria as a net effect. IgA nephropathy as well as other renal disorders present with episodes of intermittent macrohematuria that can last several days, suggesting insufficient clotting.58 One reason for this tendency of persistent bleeding can be the fibrinolytic activity of urokinase expression in the urinary tract.59

The Two Sides of the Coin—Inflammation

Exaggerated inflammation as a predominant pathomechanism

Inflammation as a major strategy of host defense often turns into the predominant pathomechanism of host defense; especially in sterile settings the downside of inflammation outweighs its minimal benefits for the host. Sterile inflammation is very common in kidney disease, eg, in autoimmune glomerulonephritis, alloimmunity of the renal transplant, renal cell necrosis, interstitial nephritis, and during tissue remodeling in progressive CKD (Figure 2a). Mononuclear phagocytes, both resident and infiltrating, express all sorts of innate pattern-recognition receptors in the kidney that translate pathogen- and cell death-related danger signals into a rapid inflammatory response.60 Mesangial cells, endothelial cells, podocytes, tubular cells, and fibroblasts express a limited number of pattern-recognition receptors and, hence, produce fewer amounts of pro-inflammatory cytokines.61 For example, renal parenchymal cells do not express TLR7 and TLR9 and, therefore, do not participate to the same extent to the immune recognition of immunostimulatory nucleic acids.62 Renal cells produce only little IL-1beta upon activation of the NLRP3 inflammasome.63, 64, 65, 66

Figure 2
figure 2

Inflammation as a pathomechanism of kidney disease. (a) Interstitial nephritis is often a sterile form of renal inflammation involving large numbers of eosinophils (black arrow) within mixed leukocyte cell infiltrates in the interstitium. Hematoxylin–eosin staining × 400. (b) Inflammation is an important mechanism of host defense during bacterial pyelonephritis. Neutrophils attack uropathogenic bacteria ascending from the urinary tract, which results in tubular white cell casts (black arrow) but also in diffuse infiltrates in the interstitium. Hematoxylin–eosin × 100.

Full size image

During sterile renal inflammation, renal cell necrosis releases endogenous ligands to pattern-recognition receptors. For example, tubular cell necrosis releases high-mobility group box 1 protein, an agonist to TLR2 and TLR4 and driver of inflammation during AKI.67, 68, 69 Tamm–Horsfall protein/uromodulin is exclusively expressed within the distal tubule and activates interstitial dendritic cells via TLR4 and NLRP3 whenever tubular injury promotes its leakage into the renal interstitium.70 Toll-like receptors and inflammasome-mediated immune recognition of necrotic cell death is a central element of rapidly progressive glomerulonephritis (RPGN), thrombotic microangiopathies, and acute tubular necrosis (ATN) as the release of TNF-alpha induces further cell necrosis, eg, necroptosis.71 Necroptosis is a caspase-independent form of cell death, highly immunogenic, that causes plasma membrane rupture and requires receptor-interacting serine-threonine kinase 3-mediated phosphorylation of the mixed lineage kinase domain-like pseudokinase. This auto-amplification loop of necrosis-related inflammation and inflammation-related necrosis was recently named 'Necroinflammation'.72 The molecular mechanisms of renal necroinflammation need to be counterbalanced to avoid widespread renal necrosis. Several molecules inhibit TLR signaling in renal immune and non-immune cells and thereby limit immune-mediated kidney diseases.73, 74, 75 Extracellular histones are another central element of necroinflammation.76 Histones are released into the extracellular space during cell necrosis where they elicit cytotoxic effects on other cells by direct charge-mediated plasma membrane disruption.77 As such, renal cell necrosis induces further histone-mediated necroinflammation.78, 79 Neutrophils undergoing NETosis are an important source of extracellular histones killing endothelial cells, eg, in septic ATN or in crescentic glomerulonephritis.80

Extrarenal infections often trigger flares of chronic glomerulonephritis or renal vasculitis but how? Circulating bacterial or viral products ligate pattern-recognition receptors in intrarenal dendritic cells and macrophages, a process triggering intrarenal production of cytokines, type I interferons that increase local inflammation and tissue damage. This process can set off renal cell loss or death, especially of podocytes.81, 82 A similar mechanism take places in lupus nephritis where the release of hypomethylated DNA or small nuclear RNA by pathogens activate dendritic cells, macrophages, and B cells.83, 84, 85 Especially the release of interferon-alpha sets off a coordinated 'pseudoantiviral' immune response, which explains why clinical manifestations of viral infections and systemic lupus are so similar.86 Also, mesangial cells and glomerular endothelial cells recognize nucleic acids and secrete type I interferons, which promotes podocyte loss and promotes glomerular scarring, eg, during viral glomerulonephritis.87, 88, 89

Intrarenal chemokine expression facilitates the sequential recruitment of various subsets of leukocytes. Thus the local microenvironment determines the functional differentiation of macrophages, T cells, and B cells into functionally distinct subsets that enforce and modulate the ongoing danger control programs.90, 91 As mentioned before, PAMPs and DAMPs turn non-activated intrarenal immune cells into effector phenotypes that confer tissue damage, eg, bacterial products induce neutrophils to undergo NETosis92 or macrophages into the M1 pro-inflammatory phenotype.93 Blocking the CC-chemokine CCL2 or its receptor CCR2 prevents the recruitment and expansion of such classically activated macrophages and thereby reduces renal immunopathology in glomeruli and the tubulointerstitium.94, 95, 96

Together, PAMPs and necrosis-related DAMPs activate renal cells to produce cytokines that drive inflammation and further regulate necrosis. Extracellular histones drive both inflammation and necrosis and leads to an auto-amplification loop of necroinflammation that largely contributes to AKI, eg, in necrotizing glomerulonephritis or ATN. Persistent renal inflammation is driven by adaptive immunity, eg, in autoimmune diseases or renal transplantation. The result is irreversible nephron loss as the central component of long-term outcomes after AKI. Therefore, suppressing renal necroinflammation is as an important therapeutic strategy to prevent nephron loss.

Insufficient inflammation in the kidney

At first, TLR-mediated renal inflammation is important for host defense during renal infection. Immune activation is absolutely necessary to suppress BK virus replication and infection in kidney allografts.97, 98 The same applies to the control of Escherichia coli and other uropathogenic bacteria during bacterial pyelonephritis99 (Figure 2b). Inflammation-mediated pathogen killing limits pathogen spreading to systemic infection, as it happens in TLR4-mutant mice. However, host defense implies collateral tissue injury such as renal abscess formation and scarring, which is absent in TLR4-mutant mice with impaired host defense.100 This inherent impaired innate immunity protects from immune-mediated kidney injury at the expense of insufficient pathogen control at the entry site and the associated risk of fatal Gram-negative sepsis. Along the same line, TLR2 recognizes leptospiral outer membrane proteins in proximal tubular epithelial cells with similar implication for leptospirosis.101

Together, evolution found a balance between the risks of overacting and insufficient inflammation as an essential danger response program. However, for some individuals either too much inflammation or insufficient inflammation becomes the predominant pathomechanism of their kidney disease.

The Two Sides of the Coin—Re-Epithelialization

Although developmentally derived from the mesenchyme, the kidneys are epithelial organs because renal functions entirely depend on the epithelial cell monolayer that covers the nephron from the glomerular tuft to the nephron segment connecting to the ureteric bud-derived collecting duct. Epithelial injuries require re-epithelialization, a process that is supported by surviving epithelial cells via cell division (proliferation=regeneration) or increasing of cell size without cell division (hypertrophy=reconstitution). Although not formally proven by lineage-tracing clonal analysis, it is widely assumed that functional recovery after ATN is associated with sufficient tubular epithelial cell regeneration.102, 103 Growth factors such as platelet-derived growth factor, EGF, HGF, and bone morphogenetic proteins induce re-epithelialization of injured epithelial monolayers.104, 105 The cell cycle regulator murine double minute-2 supports cell cycle entry of surviving tubular cells by inhibiting p53-dependent cell cycle arrest.106 Re-epithelialization requires first the resolution of inflammation because many effector elements of inflammation interfere with epithelial healing.107 The phenotype switch from 'pro-inflammatory' (M1) to 'wound healing' (M2) macrophages is important in this process. M2 macrophage-related CSF-1 secretion drives local M2 macrophage proliferation in an autocrine manner.108, 109, 110 However, a tight regulation of re-epithelialization is mandatory because overshooting as well as insufficient epithelial healing can cause disease.

Exaggerated epithelial healing as a predominant pathomechanism

When those cells with intrinsic regenerative capacity receive mitotic signals but lack those for epithelial cell differentiation, the process of regeneration can turn into a maladaptive pathomechanism and create additional renal pathology. For example, in RPGN, renal progenitors within the parietal epithelial cell (PEC) layer do no longer differentiate into podocytes but their massive hyperplasia creates cellular crescents that obstruct the Bowman's space111, 112 (Figure 3a). Crescent formation does not absolutely require inflammation because plasma leakage from injured capillaries is sufficient to drive epithelial hyperplasia, conceptually similar to the mitogenic effect of serum supplements on cultured epithelial cells in vitro.40 Indeed, in vivo glomerular epithelial cells are never exposed to serum apart from vascular injury.

Figure 3
figure 3

Exaggerated and insufficient epithelial healing in the glomerulus. (a) Podocytes are postmitotic cells that remain with us for life. Podocyte loss can lead to massive proteinuria, hence, the denudated GBM needs to be covered in either of the three ways: (1) hypertrophy of neighboring podocytes, (2) podocyte regeneration or (3) sealing of the ‘wound’ by a scar. Not infrequently, podocyte hypertrophy and regeneration are not sufficient to fully re-epithelialize the outer aspect of the glomerular capillaries; hence, focal scarring stabilizes and partially seals the defect (arrow). However, the altered hemodynamics and the associated stress on the remaining podocytes can turn insufficient epithelial healing into a pathomechanism for FSGS. PAS × 400. (b) Sometimes, podocyte loss occurs together with rupture of glomerular capillaries, eg, in crescentic glomerulonephritis. Plasma leakage is a very strong mitogenic stimulus for parietal epithelial cells, causing parietal epithelial cell hyperplasia, ie, crescent formation. Because this process destroys the glomerulus, epithelial healing becomes a pathomechanism for small vessel vasculitis. Silver staining × 400.

Full size image

Insufficient epithelial regeneration as the predominant pathomechanism

Insufficient re-epithelialization of lost glomerular podocytes is the predominant cause for CKD, its progression to end-stage kidney disease, and the spontaneous glomerulosclerosis in the aging kidney.113 The central role of podocyte loss in the pathogenesis of CKD relates to the inability of differentiated podocytes to complete mitosis for repair. Podocytes use all their cytoskeleton to maintain the secondary foot processes and the slit diaphragm along the glomerular filtration barrier, while mitosis would require simplifying cell shape and reorganizing the cytoskeleton to form the mitotic spindle. Thus podocyte loss can only be compensated by hypertrophy of surviving podocytes or by generation of de novo podocytes from podocyte progenitor cells.114 For hypertrophy, podocytes enter the S phase of the cell cycle, and cell cycle restriction points assure that they do not pass into mitosis. In case these restriction points are bypassed, podocytes detach and die, a process referred to as mitotic catastrophe.115, 116 The hypertrophy of surviving podocytes seems able to compensate a loss of up to 20–30% of podocytes; higher loss results in FSGS.117 The emergence of de novo podocytes was clearly demonstrated by lineage tracing using the podocyte-specific tomato-GFP reporter.118 However, the source of new podocytes remains under debate. Some studies suggest that bone marrow-derived progenitor cells are able to replace lost podocytes,119, 120, 121 while others propose that podocytes originate from local podocyte progenitor cells within the PECs.122, 123 Podocyte progenitors surely contribute to renal development and podocyte expansion during kidney growth in childhood but their capacity to replace lost podocytes in the adult kidney is limited. Notch and Wnt signaling, EGF, and SDF-1/CXCL12 regulate the behavior of PECs in glomerular injury.124, 125 Histone H3K9, H3K23 (acetylation), H3K4 (dimethylation), and H3K4 phosphorylation at serine 10 are associated with incomplete podocyte recovery and glomerulosclerosis.126, 127

Insufficient re-epithelialization in the tubule results in tubular atrophy. Repetitive or persistent triggers of tubule injury together with persisting classically activated macrophages impair tubular re-epithelialization after ATN. Tubular cell-derived CSF-1 enforces the local expansion of macrophages inside the kidney after injury but also drives re-epithelialization of tubuli.128 In addition, severe tubular necrosis may also eradicate tubular progenitor cells that are scattered along the tubule and that are more resistant to cell death.129 The idea that bone marrow stem cell invade the injured tubule to replace tubular cells by differentiation was experimentally excluded, but such cells still may provide paracrine support from outside the regenerating tubule.130, 131 Inability to reform the tubular epithelial monolayer leads to tubular atrophy and finally loss of the entire nephron (Figure 3b).

Together, both insufficient and exaggerated epithelial healing contributes to kidney disease. Identifying the predominant pathomechanism in individual patients and identifying specific drugs to correct the abnormal response remains a challenge.

The Two Sides of the Coin—Fibrosis/Sclerosis

Mesenchymal structures stabilize the functional units of the kidney. Mesangial cells stabilize the glomerular tuft of capillaries that confer the filtration process. Interstitial fibroblasts stabilize the tubular part of the nephron by producing interstitial matrix. In addition, mesenchymal pericytes contribute to vascular reconstruction upon injury, a process involving TIMP3 and ADAMTS1.132 Injury to the nephron or the microvasculature activates these mesenchymal elements to proliferate and produce more matrix to enforce tissue stabilization and to support the scaffold for parenchymal regeneration.1 Once parenchymal elements get irreversibly lost during the injury process, mesenchymal elements fill the space with extracellular matrix, eg, Kimmelstiel–Wilson nodules upon diabetic mesangiolysis or interstitial fibrosis upon tubular atrophy (Figure 4a).133 However, fibrosis implies stiffness (sclerosis), which may affect physiological dynamics, eg, in the glomerular tuft.134, 135

Figure 4
figure 4

Insufficient epithelial regeneration drives renal scarring, which may or may not be a pathomechanism. (a) When parietal epithelial cells in a cellular crescent undergo epithelial–mesenchymal transition, they lose their epithelial polarity and start to produce extracellular matrix around themselves, similar to mesenchymal cells, such as mesangial cells, fibroblasts, or smooth muscle cells. This creates honeycomb-like lesions in Bowman's space, which are irreversible and lead to nephron loss. Masson Trichrome staining × 400. (b) CKD is often associated with more or less confluent interstitial fibrosis together with tubular atrophy. Trichrome × 25. What is the cause and what is the effect in this association is debated. Following activation of the fibrosis danger control program mainly stabilizes the remaining nephrons and fills the spaces left behind by degenerated nephrons. This process is associated with tissue stiffness (sclerosis). An individual’s genetic background may certainly exaggerate fibrogenesis in rare cases, thus turning mesenchymal healing into a pathomechanism contributing to tubular injury. The relevance of such an exaggerated fibrosis for the progression of most CKD cases remains questionable.

Full size image

Exaggerated mesenchymal healing as a predominant pathomechanism

Mesangial regeneration upon injury can originate from several sources: surviving mesangial cells, from renin-producing cells of the extraglomerular mesangial, and from the bone marrow.136, 137, 138, 139, 140 Mesangial repair upon mesangiolysis is often studied using the rat anti-Thy1.1 model. Mesangial hyperplasia is a hallmark of so-called 'mesangioproliferative' and 'membranoproliferative glomerulonephritis', a consequence of mesangial injury in the first place.141 The stabilizing function of scars is also obvious upon insufficient podocyte regeneration. When lost podocytes cannot be replaced by PECs, focal adhesions to the Bowman’s capsule are formed. Subsequently, PECs migrate onto the glomerular tuft and lay down extracellular matrix, leading to segmental sclerosis, ie, FSGS.142 This implies that PECs not only contribute to epithelial but also to mesenchymal repair.143 Adhesions between glomerular capillaries and the Bowman’s capsule stabilize the rest of the glomerular tuft and minimize protein loss across the denudated GBM.144 However, the dynamics of hyperfiltration-related additional stress on the remaining podocytes can foster further podocyte loss and drive the progression to global glomerulosclerosis.

The extent of interstitial fibrosis correlates well with poor outcomes of primary glomerular disorders (Figure 4b). Somehow this association has led to the assumption that fibrosis could drive nephron loss and CKD progression, instead of simply being a marker of preceding nephron loss.145 There is little evidence on primary fibrotic diseases of the kidney, scleroderma does not present with renal fibrosis, and even genome-wide association studies on CKD populations did not identify risk genes pointing toward fibrogenesis.146 Nevertheless, renal fibrosis is a common finding, a process most likely driven primarily by a lack of rapid and sufficient epithelial healing, for example, tubular epithelial cells unable to regenerate get arrested in the G2/M phase and produce tumor growth factor-beta as it occurs in Chinese herb nephropathy.42 Bone marrow progenitor cells and leukocytes contribute to mesenchymal healing of the kidney as evidenced by interventions that inhibit leukocyte recruitment that can prevent directly or indirectly renal fibrogenesis.147, 148, 149, 150, 151, 152 More specifically, inhibition, deletion, or depletion of alternatively activated macrophages reduces renal fibrogenesis.153, 154, 155, 156 Ly6G+ collagen-producing 'fibrocytes' from the bone marrow invade the kidney via CCL21–CCR7 and contribute to local collagen secretion.157, 158 Also pericytes produce collagen and contribute to renal interstitial fibrosis and sclerosis.159

Insufficient mesenchymal repair in the kidney

A lack of sufficient scarring is poorly defined in renal pathology. Any form of mesangiolysis on a renal biopsy implicates an insufficient regeneration from mesangial cell progenitor cells in the extracellular mesangium.160 Persistent injury to mesangial cells and mesangial cell progenitors can be induced by massive local complement activity, eg, in atypical hemolytic uremic syndrome, immune complex glomerulonephritis, or C3 glomerulopathy.161 Also, diabetes is known to limit progenitor-driven mesangial healing. Indeed, Kimmelstiel–Wilson nodules in diabetic nephropathy are thought to result from focal mesangiolysis and secondary nodular matrix deposition.162

Together, mesenchymal repair is needed to stabilize and rebuild tissues after injury, especially after loss of parenchymal tissue. By contrast, insufficient scarring is rarely a problem in the kidney. Although being a popular concept, there is little evidence that renal fibrogenesis is really a pathogenic mechanism in the progression of CKD.163, 164, 165 Most of the quoted evidence does either not specifically target fibrogenesis (but rather inflammation or re-epithelialization) or does not report renal function as outcome parameter. It is likely that in some patients gene variants lead to overactive fibrogenesis, which could have the potential to CKD progression,166, 167 but until proven otherwise the presence of renal fibrosis in most CKD patients does not necessarily indicate it to be upstream of nephron loss.

SUMMARY AND PERSPECTIVE

Renal pathology is a cumulative expression of renal injury displaying the body's response to injury, especially in CKD. Intrinsic danger responses can be dissected into four distinct programs, ie, healing through coagulation, inflammation, re-epithelialization, and mesenchymal healing. These are all evolutionarily conserved for life-saving benefits upon traumatic injury. Each disease involves one or several of these responses to a different extent depending on the trigger mechanism. Genetic variants impairing or exaggerating these responses programs contribute to the heterogeneity of disease presentations and the predominant pathomechanism and histopathological lesion. Understanding the evolution and origin of histopathological lesions can help to avoid misconceptions in nephrology, eg, the idea that fibrosis may drive the progression of CKD, while it is nothing more than a secondary consequence of insufficient tubular regeneration and nephron loss in most patients. This concept can help to focus our research activities on how to prevent nephron loss. However, a number of questions remain and deserve further study (Table 1). Identifying the predominant pathomechanism in individual patients and identifying specific drugs to correct the abnormal tissue response remains a challenge for the future.

Table 1 Some open questions
Full size table