Glomerular endothelial cells versus podocytes as the cellular target in diabetic nephropathy
- 298 Downloads
It usually takes several years (in some cases, decades) for predisposed individuals to move from the onset of type 1 or type 2 diabetes to the development of microalbuminuria, the first sign of diabetic nephropathy. This long, complication-free, period represents the best possible moment to start a successful preventive strategy (primary prevention) aimed to avoid or at least to postpone the increase of albumin excretion rate. Prevention is based on understanding and counteracting the initial mechanisms leading to the development of the disease and unfortunately, in case of diabetic nephropathy, most of them remain unclear. Little is also known about which, among endothelial cells and podocytes, represent the first glomerular target of the complication. Selective damage of the endothelium or of the podocyte results, as a common consequence, in an increase of albumin excretion rate. Albuminuria by itself cannot therefore be of help to solve the case. Endothelium and podocytes are involved in a continuous cross-talk and by studying the impact of diabetes on this “communication” process it should be possible to obtain some information regarding the weak component of the glomerular filter. Finally, the careful investigation of the mechanisms leading to the development podocyturia, a recently identified glomerular dysfunction associated to the pathogenesis of diabetic nephropathy, could contribute to shed some more light on the very early stages of this complication.
KeywordsGlomerular endothelial cells Podocytes Diabetic nephropathy Prevention Podocyturia
Despite major efforts aimed to improve glucose metabolism and normalize blood pressure, the risk to develop end stage renal disease as a consequence of diabetic nephropathy (DN) has substantially remained stable during the last twenty years . Increased urinary excretion of albumin (the so-called microalbuminuria) represents the first clinical sign of the disease and the dysfunction that allows to discriminate between the approximately 30% of patients predisposed to develop the complication and the 70% that are instead protected, for still unclear reasons, from the disease . Despite the evidence that microalbuminuria has been discovered more than 50 years ago , and even though seminal studies [4, 5] have demonstrated a major role of insulin resistance in the pathogenesis of this diabetes-driven dysfunction, we still know very little about the pathogenic mechanisms leading to its development and which cellular components of the glomerular filtration barrier are eventually accountable for its pathogenesis. The aim of this review is to describe the morphological and functional characteristics of the glomerular filtration barrier, to discuss the multivariated effects that diabetes may exert on this structure and to explore, based on the present literature, which glomerular component might represent the specific target in case of DN.
Structure of the glomerular filter
The renal glomerulus is localized at the beginning of the nephron and consists in the actual filtering apparatus of the kidney. To be filtered and to give rise (in the Bowman space) to the so-called primary urine, the plasma has to cross a particular sieve, the glomerular filtration barrier, consisting of a fenestrated endothelium, a wide basement membrane and, finally, the slit diaphragm, an intercellular junction connecting foot processes derived from contiguous podocytes . The glomerular filtration barrier is characterized by very low permeability to macromolecules while remaining highly permeable to water and small molecules .
Fenestrations of glomerular endothelial cells are made of large, 60- to 80-nm transcellular cytoplasmic holes that are indispensable for high water permeability and urine formation . The fenestrae are at least in part covered by a protein coat known as glycocalyx  that forms a permeability barrier. The glycocalyx consists of a negatively charged network of proteoglycans, adsorbed proteins and glycosaminoglycans anchored to the cellular surface and in direct and continuous equilibrium with the plasma .
The glomerular basement membrane (GBM) is positioned between glomerular endothelial cells lying in the vascular side of the glomerular filtration barrier and podocytes that are instead situated in the urinary side . GBM is synthesized by glomerular endothelial cells and podocytes and consists of extracellular matrix and, in particular, of laminin, collagen IV, nidogen and agrin .
The final component of the glomerular filtration barrier is the slit diaphragm, a specialized structure that originates from the tight connection of foot processes derived by adjacent podocytes . Foot processes of different podocytes are connected to each other in a quite complex way through a number of different junctions that include tight, adhesion, neuronal and gap junctions, suggesting that the slit diaphragm is capable of finely regulating the plasticity of foot processes and, as a consequence, to control the filtering of macromolecules . According to recent evidences, however, the size selectivity of the slit diaphragm would be put in doubt when considering that this function could be exploited by the GBM and the podocyte glycocalyx along with the saturable tubular reabsorption of proteins .
Function of the different components of the glomerular filter
Although the obvious function of the entire glomerular filtration barrier is to filter the plasma to finally give rise to primary urine, how every single component of the barrier participate in performing this process is instead not entirely defined (Fig. 1).
The glomerular endothelium has been considered for long time a structure that permits, because of its large fenestrae, the free passage of plasma proteins. This concept has been, however, questioned after the discovery that despite the evidence that the fenestrae do not reasonably possess any kind of structural “blind” , in normal conditions albumin does not cross the glomerular endothelium and remains on its luminal side . The problem was solved after the identification of a negatively charged proteinaceous diaphragm, known as glycocalyx, covering the glomerular endothelial fenestrae . Recent evidences suggest that the effect of glycocalyx is in some way reinforced by the presence of proteins originated from the endothelial cells and forming the so-called endothelial surface layer . The sieving efficiency of glycocalyx is extremely high, micropuncture studies have in fact demonstrated that, in normal conditions, only about 0.06% of plasma albumin finally cross the glomerular capillary wall .
Although the size selectivity of the glomerular filtration barrier is largely determined by the cellular layers, the GBM has also some role in determining this parameter as it was shown that specific defects in its structure may result in the development of albuminuria .
That the slit diaphragm has a major role in controlling the permselectivity of the glomerular filtration barrier was demonstrated by studies focused on the specific slit diaphragm protein known as nephrin . In case of congenital nephrotic syndrome of the Finnish type, a disease characterized by an abnormal expression of the nephrin gene, there is in fact a massive proteinuria that is associated to loss of foot processes and slit diaphragm while in the presence of normal endothelium and GBM .
Altogether these evidences suggest that the size selectivity of the glomerular filtration barrier depends on its three layers with a reasonably major role for glomerular endothelium and slit diaphragm and, quite interestingly, the damage of every single layer may end up by causing proteinuria .
Impact of diabetes on the glomerular filter
Diabetes significantly affects the entire nephron from both a morphological and functional point of view. Increased kidney volume and faster glomerular filtration rate are indeed quite frequent, particularly in type 1 diabetic patients with short duration of the disease  and this may obviously have an effect on the filtering apparatus. Focusing on the glomerular filter, diabetes has a major impact on the entire glomerular filtration barrier.
The glomerular endothelium, just like all the different endothelia of the human bodies, is a natural target of diabetes . In particular, in case of diabetes, an increased urinary protein excretion rate in type 1 diabetic patients has been associated to a systemic reduction of the volume of the endothelial glycocalyx . Another important point comes from the demonstration that diabetes, once complicated by microalbuminuria , is associated to the loss of glomerular charge selectivity in line once again with a dysfunction of the endothelial glycocalyx. Finally, there is also the possibility that the endothelium, suffering because of persistent hyperglycemia, may end up by causing an indirect damage the podocytes . The reduced glycocalyx  that characterizes the diabetic state may in fact result in an increased delivery of macromolecules and, in particular, of proteins to the slit diaphragm, a phenomenon that immediately causes protein overload and saturation of the clearance pathways of the podocyte with consequent specific damage . In line with the hypothesis that a damaged endothelium may have a negative effect on the podocytes status during the early stages of DN, it has been recently demonstrated that in a mouse prone to develop DN, mitochondrial dysfunction of the glomerular endothelial cells is an essential feature in the pathogenesis of DN .
Moving to the GBM, we have to take into account that thickening of this structure is considered the first demonstrable sign of DN and that this dysfunction predicts the subsequent development of end stage renal disease . The reason of diabetes-driven thickening of the GBM is still not totally understood but an imbalance between synthesis and degradation of its extracellular matrix components (both functions are mainly attributed to the podocyte ) seems to have a major pathogenic role .
Finally, we have to consider the multiple effects exerted by diabetes on the podocyte. The absolute number of podocytes is significantly reduced in patients affected by DN  suggesting a direct and specific “toxic” effect of diabetes on these epithelial cells. A number of pathways are specifically activated by diabetes in the podocyte, including the ones involving advanced glycation end products (AGEs), reactive oxygen species (in particular mitochondrial oxidative stress ), cell cycle, and extra cellular matrix homeostasis . Diabetes was also shown to cause epithelial–mesenchymal transition of renal epithelial cells and in particular of podocytes, a dysfunction that would well explain the progressive loss of differentiated podocytes in case of DN . Even though the underlying mechanisms remain unclear, one important effect of chronic hyperglycemia consists in the retraction and flattening of podocyte’s foot processes, a phenomenon known as effacement that is usually followed by a significant increase of albumin excretion rate . Apoptosis is also considered a direct consequence of diabetes probably following, among other possible causes, a hyperglycemia-driven increase of reactive oxygen species in the podocytes .
Another characteristic of the podocyte that can have a specific role in case of diabetes is the fact that this cell is a recognized target of insulin . On this regard, an animal model of podocyte-specific insulin-receptor knockout (thus mimicking a case of insulin resistance) was shown to develop functional and morphologic signs of DN while remaining in a normo-glycemic state .
Glomerular endothelial cells and podocytes cross-talk in normal and high glucose
Glomerular endothelium and podocytes are involved in a continuous cross-talk. A well established way to communicate is through vascular endothelial growth factor (VEGF). VEGF is synthesized by the podocyte while glomerular endothelium expresses its receptors VEGFR-1 and VEGFR-2 . An increase of both VEGF and VEGFR-2 was shown in experimental models of DN  and, conversely, proteinuria was prevented by inhibiting VEGF or its receptors . The presence of VEGFR-1 and (although with some exceptions ) VEGFR-2 was demonstrated also in podocytes , suggesting the possibility of an autocrine regulation of these receptors.
An important factor involved in the cross-talk between glomerular endothelial cells and podocytes is also the activated protein C (APC). In case of diabetes this factor, regulated by endothelial thrombomodulin, is significantly reduced with consequent increased apoptosis of both endothelial cells and podocytes and loss of protection from the development of DN .
Exosomes (membrane-bound vesicles that can be secreted into the extracellular space) were also suggested to have a role in the cross-talk. Epithelial–mesenchymal transition of podocytes was actually shown to be mediated by exosomes derived from glomerular endothelial cells grown in high ambient glucose . Another way glomerular endothelium and podocytes get in touch is through the angiopoietins. Inside the glomerulus, Ang-1 is expressed by the podocyte  and acts on the endothelium through a specific tyrosine kinase 2 receptor (TIE2). Ang-1 supports cell survival, reduces vascular permeability and modulates the effects of VEGF . Also Ang-2, that is instead secreted only in sites of vascular remodeling , acts on Tie-2, but giving rise to opposite effects when compared to Ang-1. Ang-2 actually inhibits the binding between Ang-1 and Tie-2 . Of interest, an unbalancing of the Ang-1/Ang-2 ratio was shown to contribute to the development of DN in both animal models of diabetes  and type 2 diabetic patients . In line with these findings, it was recently shown that selective Ang-1 overexpression in podocytes of an animal model of early DN resulted in a significant (70%) reduction of albuminuria .
Beside VEGF and angiopoietins, several other ways of communication between glomerular endothelium and podocytes have been identified and include (a) stromal cell–derived factor-1 (SDF-1)/C-X-C chemokine receptor type 4 (CXCR4) axis that was suggested to be involved in the pathogenesis of glomerulosclerosis in case of type 2 diabetes . (b) semaphorins, in particular semaphorin 3A that in normal conditions is secreted by the podocyte and takes part in the glomerular development, but if overexpressed, may contribute to accelerate the progression of DN . (c) eNOS, a major player in the cross-talk between glomerular endothelium and podocytes. A tight relation has been found between eNOS deficiency and the development of DN . In line with this finding, the association between a polymorphism of the gene coding for eNOS and risk of advanced DN was also demonstrated . Finally, other possible cross-talk mediators such as endothelin-1, transforming growth factor-beta (TGF-beta), hepatocyte growth factor and insulin-like growth factor  have also been indicated as potentially involved in the pathogenesis of DN.
Altogether these findings suggest the importance of a correct cross-talk between glomerular endothelium and podocytes. Different pre-clinical/clinical trials are presently ongoing with the aim to verify the possibility to modulate this pathway. Beside the already described studies based on the inhibition of VEGF activity , exogenous administration of activated protein C  and inhibition of epidermal growth factor receptor tyrosine kinase activity  were shown to have beneficial effects on DN in experimental models of diabetes. Finally, even though not yet tested directly on DN, the Wnt antagonist DKK2 (that modulates the angiopoietin-1-Tie2 pathway) was found to be very effective in case of diabetes .
In perspective more information concerning, the way endothelial cells and podocytes interact will reasonably come from new technologies that include advanced electron microscopic techniques such as the focused ion beam-scanning electron microscopy (FIB-SEM)  and by the possibility to create a “glomerulus-on-a-chip” taking advantage of induced pluripotent stem cell-derived human podocytes co-cultured with human glomerular endothelial cells .
Finally, it has to be considered that the cross-talk between endothelium and podocytes can also be influenced by other glomerular components such as the mesangial cells that, particularly in case of diabetes, may end up by causing mesangial expansion  and the parietal epithelial cells that, being able to differentiate into podocytes, have a potential indirect impact on the endothelium–podocytes cross-talk .
Endothelial cells and podocytes: which is the first target of diabetes?
Although the above described evidences indicate that both glomerular endothelial cells and podocytes are involved, either directly or indirectly in the pathogenesis of DN, the question concerning which is the first cellular target of diabetes remains unsolved.
A potential new source of information in the search for the cellular target of diabetes: the emergence of podocyturia
DN is characterized by a progressive reduction of the number of podocytes inside the glomerulus . Death of podocytes and in particular apoptosis (preceded by podocyte dysfunction and consequent reduced expression of podocyte markers ), but also insufficient podocyte autophagy  are considered the major causes of this phenomenon. Recent studies suggest, however, that the podocyte is rather resistant to apoptosis and is able to maintain viability even in extreme situations. With the aim to remain viable, the podocyte progressively loses the interdigitating foot process pattern, a phenomenon known as foot process effacement . This defensive conduct allows the surviving of the podocyte, but at the same time, makes this cell susceptible to detachment . Once detached from the GBM the podocyte ends up in the Bowman’s space and is then rapidly excreted, still viable, with urine. This phenomenon, known as urinary excretion of podocytes or podocyturia  is present, in small amount, in healthy individuals [64, 65] and, to a higher extent, in case of DN . The functional and molecular characterization of the physiologic and pathologic forms of podocyturia could represent a key finding to clarify the mechanisms leading to podocyte detachment in case of chronic renal diseases and in particular of DN.
Detachment of podocytes from the GBM is a well established feature of DN  but it is still unclear whether this dysfunction is the result of an “aggression” of diabetes directly focused on the podocyte or if it is instead the consequence of a more canonical glucotoxicity aimed at the glomerular endothelium.
Of interest, when kidney biopsies from type 2 diabetic patients with and without DN were compared, ultrastructural changes of podocytes were already observed at the stage of microalbuminuria . This finding is consistent with a direct podocyte injury characterized by foot process effacement and pseudocysts as can be seen at scanning electron microscopy , a scenario potentially compatible with the pathogenesis of podocyturia . Conversely, in case of type 1 diabetes, although foot process width and podocyte detachment were once again increased in nephropatic patients, reduced endothelial cell fenestration was already detectable in normoalbuminuric patients, possibly suggesting that in this case, endothelial dysfunction might precede the damage to the podocyte .
From all these data, it is quite evident that glomerular endothelial cells and podocytes are both necessary for the correct functioning of the glomerular filtration barrier and that the barrier reasonably works properly only when all its components are in optimal conditions. When damaged, both glomerular endothelial cells and podocytes may give rise to albuminuria so it is not easy to clarify which cell represents the first target of diabetes particularly when considering that albuminuria represents the first sign of DN. Studies aimed to expand our functional and molecular knowledge of the different components of the glomerular filtration barrier on one hand, and to clarify, on the other hand, the pathogenesis of specific dysfunctions such as podocyturia will be of great help to clarify the time course of the impact of diabetes on the glomerular filter, a key step to set up a specific strategy aimed to successfully prevent the development of DN.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants or animals performed by any of the authors.
Informed consent is not required because no studies in humans were performed.
- 4.Parvanova AI, Trevisan R, Iliev IP, Dimitrov BD, Vedovato M, Tiengo A et al (2006) Insulin resistance and microalbuminuria: a cross-sectional, case-control study of 158 patients with type 2 diabetes and different degrees of urinary albumin excretion. Diabetes 55(5):1456–1462. https://doi.org/10.2337/db05-1484 CrossRefPubMedGoogle Scholar
- 5.De Cosmo S, Motterlini N, Prudente S, Pellegrini F, Trevisan R, Bossi A et al (2009) Impact of the PPAR-gamma2 Pro12Ala polymorphism and ACE inhibitor therapy on new-onset microalbuminuria in type 2 diabetes: evidence from BENEDICT. Diabetes 58(12):2920–2929. https://doi.org/10.2337/db09-0407 CrossRefPubMedPubMedCentralGoogle Scholar
- 17.Holthofer H, Ahola H, Solin ML, Wang S, Palmen T, Luimula P et al (1999) Nephrin localizes at the podocyte filtration slit area and is characteristically spliced in the human kidney. Am J Pathol 155(5):1681–1687. https://doi.org/10.1016/S0002-9440(10)65483-1 CrossRefPubMedPubMedCentralGoogle Scholar
- 20.Zerbini G, Bonfanti R, Meschi F, Bognetti E, Paesano PL, Gianolli L et al (2006) Persistent renal hypertrophy and faster decline of glomerular filtration rate precede the development of microalbuminuria in type 1 diabetes. Diabetes 55(9):2620–2625. https://doi.org/10.2337/db06-0592 CrossRefPubMedGoogle Scholar
- 25.Morigi M, Buelli S, Angioletti S, Zanchi C, Longaretti L, Zoja C et al (2005) In response to protein load podocytes reorganize cytoskeleton and modulate endothelin–1 gene: implication for permselective dysfunction of chronic nephropathies. Am J Pathol 166(5):1309–1320. https://doi.org/10.1016/S0002-9440(10)62350-4 CrossRefPubMedPubMedCentralGoogle Scholar
- 31.Sagoo MK, Gnudi L (2018) Diabetic nephropathy: Is there a role for oxidative stress? Free Radic Biol Med 116:50–63. https://doi.org/10.1016/j.freeradbiomed.2017.12.040 CrossRefPubMedGoogle Scholar
- 34.Salmon AH, Toma I, Sipos A, Muston PR, Harper SJ, Bates DO et al(2007) Evidence for restriction of fluid and solute movement across the glomerular capillary wall by the subpodocytes pace. Am J Physiol Renal Physiol 293(6):F1777–F1786. https://doi.org/10.1152/ajprenal.00187.2007 CrossRefPubMedGoogle Scholar
- 48.Lim HS, Lip GY, Blann AD (2005) Angiopoietin-1 and angiopoietin-2 in diabetes mellitus: relationship to VEGF, glycaemic control, endothelial damage/dysfunction and atherosclerosis. Atherosclerosis 180(1):113–118. https://doi.org/10.1016/j.atherosclerosis.2004.11.004 CrossRefPubMedGoogle Scholar
- 49.Sayyed SG1, Hägele H, Kulkarni OP, Endlich K, Segerer S, Eulberg D et al (2009) Podocytes produce homeostatic chemokine stromal cell-derived factor-1/CXCL12, which contributes to glomerulosclerosis, podocyte loss and albuminuria in a mouse model of type 2 diabetes. Diabetologia 52(11):2445–2454. https://doi.org/10.1007/s00125-009-1493-6 CrossRefPubMedGoogle Scholar
- 52.Zanchi A, Moczulski DK, Hanna LS, Wantman M, Warram JH, Krolewski AS (2000) Risk of advanced diabetic nephropathy in type 1 diabetes is associated with endothelial nitric oxide synthase gene polymorphism. Kidney Int 57(2):405–413. https://doi.org/10.1046/j.1523-1755.2000.00860.x CrossRefPubMedGoogle Scholar
- 54.Zhang M-Z, Wang Y, Paueksakon P, Harris RC (2014) Epidermal growth factor receptor inhibition slows progression of diabetic nephropathy in association with a decrease in endoplasmic reticulum stress and an increase in autophagy. Diabetes 63(6):2063–2072. https://doi.org/10.2337/db13-1279 CrossRefPubMedPubMedCentralGoogle Scholar
- 56.Burghardt T, Hochapfel F, Salecker B, Meese C, Gröne HJ, Rachel R et al (2015) Advanced electron microscopic techniques provide a deeper insight into the peculiar features of podocytes. Am J Physiol Renal Physiol 309(12):F1082–F1089. https://doi.org/10.1152/ajprenal.00338.2015 CrossRefPubMedGoogle Scholar
- 68.Weil EJ, Lemley KV, Mason CC, Yee B, Jones LI, Blouch K et al (2012) Podocyte detachment and reduced glomerular capillary endothelial fenestration promote kidney disease in type 2 diabetic nephropathy. Kidney Int 82(9):1010–1017. https://doi.org/10.1038/ki.2012.234 CrossRefPubMedPubMedCentralGoogle Scholar