sign in sign up
<<
Home , Nephrotic syndrome, Proteinuria

BRIEF REVIEW www.jasn.org

Nephrotic Syndrome: Components, Connections, and Angiopoietin-Like 4–Related Therapeutics

Camille Macé and Sumant S. Chugh

Glomerular Disease Therapeutics Laboratory, University of Alabama at Birmingham, Birmingham, Alabama

ABSTRACT is recognized by the presence of in excess of 3.5 g/24 for FSGS4). These genes are h along with , , ( and best viewed as disease-related genes, ), and lipiduria. Each component has been investigated individ- with proteinuria being a major compo- ually over the past four decades with some success. Studies published recently have nent of the associated disease. Others started unraveling the molecular basis of proteinuria and its relationship with other were discovered through mutagenesis components. We now have improved understanding of the threshold for nephrotic- studies in mice (e.g., neph1).5,6 In most range proteinuria and the pathogenesis of hypertriglyceridemia. These studies reveal cases, these genes do not single-handedly that modifying sialylation of the soluble glycoprotein angiopoietin-like 4 or changing explain the development of proteinuria. key amino acids in its sequence can be used successfully to treat proteinuria. Treatment Transcriptional factors, like WT1, came strategies on the basis of fundamental relationships among different components of into light from a combination of genet- nephrotic syndrome use naturally occurring pathways and have great potential for ics7 and animal model8 studies, whereas future development into clinically relevant therapeutic agents. the role of ZHX was discovered through differential gene expression J Am Soc Nephrol 25: 2393–2398, 2014. doi: 10.1681/ASN.2014030267 studies in animal models.2,9 Among po- docyte-secreted proteins implicated in human disease, the roles of angiopoietin- Nephrotic syndrome is a hallmark of glo- about each component and then consider like 4 (Angptl4) and vascular endothelial merular disease and characterized by the what remains to be investigated. Next, mo- growth factor in proteinuria were estab- presence of proteinuria in excess of 3.5 g/24 lecular connections, if known, between lished through the study of experimental h, hypoalbuminemia, and variable amounts proteinuria and the other components models.10–12 A hyposialylated form of of hyperlipidemia (hypertriglyceridemia will be discussed. Finally, novel therapeutic Angptl4 secreted from is di- and hypercholesterolemia), lipiduria, and strategies derived from the study of ne- rectly implicated in the pathogenesis edema1 (Figure 1). In children, nephrotic- phrotic syndrome will be outlined. of proteinuria in MCD and accounts range proteinuria is defined by urinary for most of the cardinal manifestations excretion rates .40 mg/h per of this disease, including glucocorticoid meter2. Patients with primary glomeru- PROTEINURIA sensitivity, selective proteinuria, loss lar diseases (e.g., of glomerular basement membrane [MCD], FSGS, and membranous ne- Several genes expressed in podocytes (GBM) charge, and classic morphologic phropathy) and systemic disorders (e.g., have now been directly or indirectly changes.1,11 Despite demonstration that mellitus, systemic erythe- implicated in the pathogenesis of pro- loss of GBM charge in MCD is caused by matosis, and ) can present teinuria. They can be classified as slit binding of Angptl4, it is not known with nephrotic syndrome. The principle diaphragm-related, cell matrix interface– driving force in nephrotic syndrome is related, cytoskeleton-related, Published online ahead of print. Publication date proteinuria, because other components surface proteins, transcriptional factors, available at www.jasn.org. develop only after proteinuria reaches a and podocyte-secreted proteins.2 Many certain threshold. Substantial research ef- of the structural proteins among these Correspondence: Dr.SumantS.Chugh,University of Alabama at Birmingham Division of , fort has been committed to understanding categories were discovered when screen- THT 611L, 1900 University Boulevard, Birmingham, the pathogenesis of each individual com- ing for mutations in patients with disease AL 35294. Email: [email protected] ponent. The purpose of this review is to (e.g., nephrin for congenital nephrotic Copyright © 2014 by the American Society of discussinbroadtermswhatisunderstood syndrome of the Finnish type3 and Nephrology

J Am Soc Nephrol 25: 2393–2398, 2014 ISSN : 1046-6673/2511-2393 2393 BRIEF REVIEW www.jasn.org

efficacy, transporters in the thick ascending loop of Henle, distal tubule, collecting duct, and have been targeted to reduce edema. Re- cent studies suggest that the proteolytic ac- tivation of collecting duct epithelial channel may be mediated by plasmin con- verted from filtered plasminogen.17,18 Treatment of edema with infu- sions is generally not practiced, except in selected cases of refractory . Perhaps the least understood part of edema is increased peripheral capillary permeability, and no current therapy targets this aspect. This area has tremen- dous potential for investigation, and mechanistic studies may reveal useful clues for treating patients with refrac- tory edema. Investigating the explosive onset of edema in MCD could provide insight into additional molecular mech- anisms in nephrotic syndrome, because Figure 1. The nephrotic syndrome tree is shown. The trunk depicts increasing proteinuria, there is a potential for glomerular or pe- and the branches represent other components that appear when proteinuria crosses the ripherally secreted proteins in the path- nephrotic-range threshold. ogenesis of this phenomenon. whether complex interaction of this pro- secreting proteins into the circulation. HYPERCHOLESTEROLEMIA tein with heparan sulfate proteoglycans Angptl4 is the first of several such pro- and other GBM proteins or actual loss teins that are likely to be identified in the Nephrotic patients have elevated total of GBM charge is responsible for protein- future. andLDL levels, largely related uria. Vascular endothelial growth factor to an acquired LDL receptor deficiency, has been implicated in human throm- which limits the removal of cholesterol- botic microangiopathy.13 The added di- EDEMA rich LDL particles from the circulation.19 mension of podocyte-secreted proteins is This reduction in uptake of extracellu- their ability to reach out to binding part- The onset of edema in nephrotic syn- lar cholesterol stimulates cholesterol ners on the surface of glomerular endo- drome occupies a clinical spectrum that biosynthesis by upregulating hepatic thelial cells and potentially participate in varies from subacute to onset in 3-hydroxy-3-methyl glutaryl-CoA reduc- feedback loops within the .14 many patients with FSGS or membranous tase expression and activity in the ne- In addition to podocyte-secreted pro- nephropathy, explosive onset (often over- phrotic . Increased hepatic activity teins, high plasma levels of the soluble night) in MCD, and complete absence in of Acyl-CoA cholesterol acyltransferase- urokinase receptor are being investigated many patientswithHIV-related collapsing 2, an enzyme responsible for esterifica- for their role in the pathogenesis of . From a pathophysiology tion of cholesterol, is also noted. Presence FSGS.15 standpoint, edemarequires a combination of normal LDL receptor mRNA expres- A new dimension recently added to of hypoalbuminemia, renal salt retention, sion in the nephrotic liver suggests a this field is the systemic response to and increased peripheral capillary perme- post-transcriptional etiology. A recent proteinuria when it reaches nephrotic ability,becausetherearenumerousclinical study in experimental animals suggests range. A circulating sialylated form of situations involving a single component that increased hepatic degradation of Angptl4 secreted predominantly from that are not associated with edema. The the LDL receptor by proprotein convertase adipose tissue, skeletal muscle, and heart variability of edema in different clinical subtilisin/kexin type 9 and inducible de- reduces proteinuria, at least in part, by situations may be directly related to differ- grader of the LDL receptor may be in- 20 binding to glomerular endothelial avb5 ences between these pathogenic compo- volved. Also, urinary loss of plasma pro- integrin.12 Thematically, it opens up a new nents. The best studied aspect and indeed, teins like lecithin–cholesterol acyltransferase area of investigation to study how other the primary target of diuretic therapy is may also contribute to hypercholesterol- organs reduce established proteinuria by renal tubular salt retention.16 In order of emia. However, the precise sequence of

2394 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 2393–2398, 2014 www.jasn.org BRIEF REVIEW events and the molecular relationship of resulting in the accumulation of albu- MOLECULAR LINKS BETWEEN these changes with proteinuria remain min with higher FFA content.12,22,23 PROTEINURIA AND OTHER unknown. This loss, along with the development COMPONENTS OF NEPHROTIC of hypoalbuminemia, results in a high SYNDROME plasma FFA-to-albumin ratio, which in HYPERTRIGLYCERIDEMIA turn, drives FFA uptake in skeletal mus- There are large gaps of knowledge in our cle, heart, and adipose tissue.12 This understanding of the molecular relation- Hypertriglyceridemia results from im- event is primarily responsible for the de- ship between proteinuria, the primary paired clearance of in very velopment of hypertriglyceridemia and driver in nephrotic syndrome, and most low-density and chylomicrons as discussed below, an attempt of these of the other components. Only the link because of inactivation of endothelium- organs to reduce proteinuria through between proteinuria and hypertriglycer- bound (LPL) activity by the secretion of Angptl4 into the circu- idemia has been clearly elucidated.12,14 the circulating glycoprotein Angptl4, lation. This relationship is strongly influenced which reduces the conversion of circulat- by the link between FFA and albumin. ing triglycerides to free fatty acids (FFAs).12 FFAs are a major fuel source for skeletal Circulating sialylated Angptl4 is mostly LIPIDURIA muscle and heart, and they are stored in secreted from skeletal muscle, adipose adipose tissue as triglycerides. Under tissue, and heart to reduce proteinuria Lipiduria is thought to be secondary to normal conditions, FFA uptake in these through glomerular endothelial binding, hyperlipidemia, and it mostly results organs relies on a combination of FFA but it also causes hypertriglyceridemia from filtration of HDL particles because released by LPL-mediated hydrolysis of as a side effect. Interaction of Angptl4 of their relatively small size.24 These lip- circulating and albumin- with LPL converts the active dimeric ids are found in oval fat bodies (sloughed bound FFAs that are derived from form of this protein into inactive mon- tubular cells with lipids), fatty casts, or (medium-chain FFA) or lipolysis of omers. Both dimers and monomers of free-floating lipid globules. Components stored triglyceride in adipose tissue in LPL are lost in the in nephrotic that have high amounts of esterified cho- the fasting state (Figure 2). These organs syndrome. Proteinuria and hypertri- lesterol have a Maltese cross appearance also have high expression for LPL, glyceridemia are linked by two negative under polarized light. Angptl4, and peroxisome proliferator- feedback loops, which are discussed below.

HYPOALBUMINEMIA

Hypoalbuminemia results from urinary losses of albumin during proteinuria, insufficientcompensationbyhepatic synthesis, and perhaps, increased albu- min catabolism. The major enigma in the pathogenesis of hypoalbuminemia is the inability of the nephrotic liver to increase albumin synthesis to compensate for urinary losses, although a normal liver synthesizes 12–14 g albumin/d and can increase production 3-fold in times of demand.21 Whereas a lot of importance was traditionally placed on reduced plasma resulting from hypoalbuminemia, it has now become clear that changes in FFA binding to al- Figure 2. Schematic illustration of the two sources of FFA available for uptake by skeletal 12 bumin are equally important. Plasma muscle, heart, and adipose tissue in the normal and nephrotic state. Green shows normal FFAs are noncovalently bound to albu- conditions, and red illustrates changes in nephrotic syndrome. The balance shifts signifi- fi min through six high-af nity sites and cantly to albumin-bound FFA because of retention of albumin with high FFA content in several low-affinity sites.14 During pro- nephrotic syndrome. Angptl4 secreted from these organs reduces the conversion of tri- teinuria, patients lose, for unclear rea- glycerides to FFA by inactivating LPL, thereby reducing use of triglycerides and resulting in sons, albumin with lower FFA content, hypertriglyceridemia.

J Am Soc Nephrol 25: 2393–2398, 2014 Components of Nephrotic Syndrome 2395 BRIEF REVIEW www.jasn.org activated receptor (PPAR) family mem- bers, which regulate Angptl4 expression in response to FFA uptake. In nephrotic syndrome, there is urinary loss of al- bumin with low FFA content, leading to retention of albumin with high FFA content, which alters the FFA uptake bal- ance in favor of albumin-bound FFA. In- crease in uptake of albumin-bound FFA induces local Angptl4 upregulation, likely through PPARs, which then inac- tivates LPL in the same tissues, leading to reduction in the formation of FFA from triglycerides and consequently, hypertri- glyceridemia. This change in balance be- tween two sources of FFA can be viewed as a product of a local negative feedback loop (Figure 3). However, this local feed- back loop is subservient to a much larger systemic negative feedback loop to re- duce proteinuria. Indeed, Angptl4 re- leased from the skeletal muscle, heart, and adipose tissue into the circulation also binds to glomerular endothelial avb5 integrin to reduce proteinuria. Pa- tients with all causes of nephrotic syn- drome studied have elevated plasma Angptl4 levels.12 Studies in diabetic rats using low doses of recombinant human Angptl4 suggest a lower threshold to par- ticipate in the systemic rather than the lo- cal feedback loop. Overall, it seems that the local feedback loop reduces the effec- tiveness of the systemic feedback loop by limiting the extent of Angptl4 upregula- tion. Overall, the antiproteinuric and hy- pertriglyceridemia-inducing effects of Angptl4 are dependent on high uptake of albumin-bound FFA in peripheral or- gans and indirectly dependent on dispro- portionate retention of albumin with high FFA content in nephrotic syndrome.

MOLECULAR BASIS FOR NEPHROTIC-RANGE Figure 3. Schematic illustration of negative feedback loops in the link between proteinuria, PROTEINURIA THRESHOLD hypoalbuminemia, and hypertriglyceridemia mediated by Angptl4 and FFA. Plasma FFAs are noncovalently bound to albumin. Because of the preferential loss of albumin with low Patients do not start manifesting the FFA content in nephrotic syndrome, there is a relative increase in circulating albumin with other components of nephrotic syn- higher FFA content. Because glomerular disease progresses to severe proteinuria, hypo- drome until they cross the nephrotic- albuminemia develops, and the combination of high albumin FFA content and lower plasma range proteinuria threshold, which is albumin levels increases the FFA-to-albumin ratio. It promotes entry of FFA into skeletal muscle, heart, and adipose tissue, which causes upregulation of Angptl4 at least partially usually defined as about 3.5 g/d in adults, mediated by PPARs. Angptl4 secreted from these organs participates in two feedback but it is likely to be quite variable between loops. In the systemic loop, it binds to glomerular endothelial avb5 integrin and reduces different individuals and within the same

2396 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 2393–2398, 2014 www.jasn.org BRIEF REVIEW individual among different components. systemic feedback loop. However, these This area of investigation would benefit Until recently, the molecular basis for interventions run the risk of aggravating from additional clinical studies in the this threshold was not known. It is, at hypertriglyceridemia through the local future. Perhaps other unresolved or par- present, only possible to explain the feedback loop and could also reduce tially resolved components of nephrotic nephrotic threshold in the context of FFAuptakeintheheartandskeletal syndrome may also result from a similar hypertriglyceridemia.12 Hypertriglyceri- muscle to below a threshold for organ systemic response involving other puta- demia is dependent on the retention of dysfunction. In addition, all the tive proteins. Identifying and modifying albumin with high FFA content, result- mentioned above have multiple side ef- this circulating glomerulophilic pro- ing in a plasma FFA-to-albumin ratio fects. Mutating Angptl4 at amino acid 40 teome may hold the key to developing that is high enough to induce upregula- or 39 to reduce its interaction with LPL additional novel therapies for proteinuric tion of Angptl4 expression in skeletal bypasses the local feedback loop and al- disease in the future. muscle, heart, and adipose tissue. Stud- lows for mutant recombinant human ies in experimental animals reveal that, Angptl4 to very significantly reduce pro- during mild proteinuria, plasma FFA-to- teinuria in nephrotic animals with FSGS albumin ratio, plasma Angptl4 levels, or without affect- ACKNOWLEDGMENTS peripheral organ Angptl4, and PPAR ing plasma triglyceride levels.12 This mRNA expression are similar to control strategy is a novel futuristic treatment This work was supported by National In- nonproteinuric animals. During severe for all etiologies of proteinuria and ne- stitutes of Health Grants T32-DK007545 (to proteinuria, all these parameters are phrotic syndrome, especially if studies C.M.), R01-DK077073 (to S.S.C.), R01- significantly elevated, suggesting that on the long-term administration of DK090035 (to S.S.C.), and R01-DK101637 the threshold for nephrotic-range pro- mutant human Angptl4 currently in (to S.S.C.). teinuria, in the context of hypertriglycer- progress also show improvement in the idemia, correlates with the downstream progression of CKD. effects of an increased plasma FFA-to- It is also clear that podocyte-secreted DISCLOSURES albumin ratio. hyposialylated Angptl4 mediates pro- S.S.C. is Founder, President, and Chief Executive fi fi teinuria in MCD1,11 and also contributes Of cer of GDTHERAPY LLC and led patents to proteinuria in diabetic nephropathy.14 related to the use of Angptl4 mutants (PCT/ US2011/039255) and precursors of sialic acid, THERAPEUTIC STRATEGIES The effects of hyposialylated Angptl4 are including N-acetyl-D-manosamine (PCT/US2011/ DEVELOPED FROM THE STUDY OF most likely related to its binding to the 039058), for the treatment of nephrotic syndrome. 11 NEPHROTIC SYNDROME GBM, but adverse consequences of glo- He may benefit financially from these patents in the merular endothelial binding are also future. C.M. declares no competing financial inter- A central role played by Angptl4 in possible.12 Converting hyposialylated ests. nephrotic syndrome makes it suitable Angptl4 to sialylated protein using as a therapeutic agent as well as a tar- N-acetyl-D-manosamine, a precursor of get.14,25 An important lesson learned sialic acid that can be taken up and stored REFERENCES from Angptl4 is its “Jekyll and Hyde” ef- in podocytes, very significantly reduces fect, because the same protein can have proteinuria and has the potential for 1. Chugh SS, Clement LC, Macé C: New in- sights into human minimal change disease: very different biologic effects on the ba- use in small maintenance doses to pre- Lessons from animal models. Am J Kidney sisofthepresenceorabsenceofsialic vent relapse in MCD and as maintenance Dis 59: 284–292, 2012 acid residues and the compartment in therapy for diabetic nephropathy.14 2. Clement LC, Liu G, Perez-Torres I, Kanwar YS, which it is secreted. Because circulating Avila-Casado C, Chugh SS: Early changes in fl sialylated Angptl4 has an antiproteinuric gene expression that in uencethecourseof primary glomerular disease. Kidney Int 72: effect, it is conceivable that injecting re- FUTURE INTERVENTIONS ON THE 337–347, 2007 combinant human Angptl4 or increas- BASIS OF THE STUDY OF 3. Kestilä M, Lenkkeri U, Männikkö M, Lamerdin ing the secretion of native Angptl4 NEPHROTIC SYNDROME J, McCready P, Putaala H, Ruotsalainen V, from heart, skeletal muscle, and adipose Morita T, Nissinen M, Herva R, Kashtan CE, tissue using PPAR agonists, glucocorti- The proteinuria–hypertriglyceridemia Peltonen L, Holmberg C, Olsen A, Tryggvason b K: Positionally cloned gene for a novel glomer- coids, FFA supplements, or -adrenergic relationship shows that other organs in ular protein—nephrin—is mutated in congen- agonists could be of potential benefitin the body make an attempt to reduce pro- ital nephrotic syndrome. Mol Cell 1: 575–582, reducing proteinuria through the teinuria by secreting specificproteins. 1998 4. Boute N, Gribouval O, Roselli S, Benessy F, Lee H, Fuchshuber A, Dahan K, Gubler MC, Niaudet P, Antignac C: NPHS2, encoding the proteinuria. In a local loop, it inactivates LPL in the same organs from which it is secreted to glomerular protein podocin, is mutated in au- reduce the uptake of FFA, thereby curtailing a stimulus for its own upregulation. Repro- tosomal recessive -resistant nephrotic duced from reference 14, with permission. syndrome. Nat Genet 24: 349–354, 2000

J Am Soc Nephrol 25: 2393–2398, 2014 Components of Nephrotic Syndrome 2397 BRIEF REVIEW www.jasn.org

5. Donoviel DB, Freed DD, Vogel H, Potter DG, 11. Clement LC, Avila-Casado C, Macé C, Soria activates epithelial Na+ channels by cleaving Hawkins E, Barrish JP, Mathur BN, Turner CA, E, Bakker WW, Kersten S, Chugh SS: Podo- the gamma subunit. J Biol Chem 283: Geske R, Montgomery CA, Starbuck M, cyte-secreted angiopoietin-like-4 mediates 36586–36591, 2008 Brandt M, Gupta A, Ramirez-Solis R, proteinuria in glucocorticoid-sensitive ne- 18. Svenningsen P, Bistrup C, Friis UG, Bertog M, Zambrowicz BP, Powell DR: Proteinuria and phrotic syndrome. Nat Med 17: 117–122, Haerteis S, Krueger B, Stubbe J, Jensen ON, perinatal lethality in mice lacking NEPH1, a 2011 Thiesson HC, Uhrenholt TR, Jespersen B, novel protein with homology to NEPHRIN. 12.ClementLC,MacéC,Avila-CasadoC,Joles Jensen BL, Korbmacher C, Skøtt O: Plasmin Mol Cell Biol 21: 4829–4836, 2001 JA, Kersten S, Chugh SS: Circulating angio- in nephrotic urine activates the epithelial 6. Liu G, Kaw B, Kurfis J, Rahmanuddin S, poietin-like 4 links proteinuria with hyper- sodium channel. JAmSocNephrol20: 299– Kanwar YS, Chugh SS: Neph1 and nephrin triglyceridemia in nephrotic syndrome. Nat 310, 2009 interaction in the slit diaphragm is an impor- Med 20: 37–46, 2014 19. Vaziri ND: Molecular mechanisms of lipid tant determinant of glomerular permeability. 13. Eremina V, Jefferson JA, Kowalewska J, disorders in nephrotic syndrome. Kidney Int JClinInvest112: 209–221, 2003 Hochster H, Haas M, Weisstuch J, Richardson 63: 1964–1976, 2003 7. Pelletier J, Bruening W, Kashtan CE, Mauer C, Kopp JB, Kabir MG, Backx PH, Gerber HP, 20. Jin K, Park BS, Kim YW, Vaziri ND: Plasma SM, Manivel JC, Striegel JE, Houghton DC, Ferrara N, Barisoni L, Alpers CE, Quaggin SE: PCSK9 in nephrotic syndrome and in perito- Junien C, Habib R, Fouser L, Fine RN, VEGF inhibition and renal thrombotic mi- neal : A cross-sectional study. Am J Silverman BL, Haber DA, Housman D: croangiopathy. NEnglJMed358: 1129– Kidney Dis 63: 584–589, 2014 Germline mutations in the Wilms’ tumor 1136, 2008 21. Rothschild MA, Oratz M, Schreiber SS: Al- suppressor gene are associated with abnor- 14. Chugh SS, Macé C, Clement LC, Del Nogal bumin synthesis. 1. N Engl J Med 286: 748– mal urogenital development in Denys-Drash Avila M, Marshall CB: Angiopoietin-like 4 757, 1972 syndrome. Cell 67: 437–447, 1991 based therapeutics for proteinuria and kid- 22. Ghiggeri GM, Ginevri F, Candiano G, 8. Barisoni L, Bruggeman LA, Mundel P, ney disease. Front Pharmacol 5: 23, 2014 Oleggini R, Perfumo F, Queirolo C, Gusmano D’Agati VD, Klotman PE: HIV-1 induces renal 15. Wei C, El Hindi S, Li J, Fornoni A, Goes N, R: Characterization of cationic albumin in epithelial dedifferentiation in a transgenic Sageshima J, Maiguel D, Karumanchi SA, minimal change nephropathy. Kidney Int 32: model of HIV-associated nephropathy. Kid- Yap HK, Saleem M, Zhang Q, Nikolic B, 547–553, 1987 ney Int 58: 173–181, 2000 Chaudhuri A, Daftarian P, Salido E, Torres A, 23. Lerique B, Moulin B, Delpero C, Purgus R, 9. Liu G, Clement LC, Kanwar YS, Avila-Casado Salifu M, Sarwal MM, Schaefer F, Morath C, Olmer M, Boyer J: High-affinity interaction of C, Chugh SS: ZHX proteins regulate podo- Schwenger V, Zeier M, Gupta V, Roth D, long-chain fatty acids with albumin in cyte gene expression during the de- Rastaldi MP, Burke G, Ruiz P, Reiser J: Cir- nephrotic syndrome. Clin Sci (Lond) 89: 417– velopment of nephrotic syndrome. J Biol culating urokinase receptor as a cause of fo- 420, 1995 Chem 281: 39681–39692, 2006 cal segmental . Nat Med 24. Blackburn V, Grignani S, Fogazzi GB: Lip- 10. Eremina V, Sood M, Haigh J, Nagy A, Lajoie 17: 952–960, 2011 iduria as seen by transmission electron mi- G, Ferrara N, Gerber HP, Kikkawa Y, Miner 16. Hamm LL, Batuman V: Edema in the nephrotic croscopy. Nephrol Dial Transplant 13: 2682– JH, Quaggin SE: Glomerular-specificalter- syndrome: New aspect of an old enigma. JAm 2684, 1998 ations of VEGF-A expression lead to distinct Soc Nephrol 14: 3288–3289, 2003 25. Kirk R: Nephrotic syndrome: Negative feed- congenital and acquired renal diseases. J 17. Passero CJ, Mueller GM, Rondon-Berrios H, back loop reveals novel potential therapy. Clin Invest 111: 707–716, 2003 Tofovic SP, Hughey RP, Kleyman TR: Plasmin Nat Rev Nephrol 10: 63, 2014

2398 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 2393–2398, 2014