Renal Physiology

The Glomerulus: The Sphere of Influence

Martin R. Pollak,* Susan E. Quaggin,† Melanie P. Hoenig,* and Lance D. Dworkin‡

AbstractThe glomerulus, the filtering unit of the kidney, is a unique bundle of capillaries lined by delicate fenestratedendothelia, a complex mesh of proteins that serve as the glomerular basement membrane and specializedvisceral epithelial cells that form the slit diaphragms between interdigitating foot processes. Taken together,this arrangement allows continuous filtration of the plasma volume. The dynamic physical forces thatdetermine the single nephron glomerular filtration are considered. In addition, new insights into the cellularand molecular components of the glomerular tuft and their contribution to glomerular disorders areexplored.

Clin J Am Soc Nephrol 9: 1461–1469, 2014. doi: 10.2215/CJN.09400913

The GlomerulusThe glomerulus, the filtering unit of the kidney, is aspecialized bundle of capillaries that are uniquelysituated between two resistance vessels (Figure 1).These capillaries are each contained within theBowman’s capsule and they are the only capillarybeds in the body that are not surrounded by intersti-tial tissue. Therefore, a unique support structure isneeded to maintain flow in these essential capillaryunits. In fact, all of the major components of the filteritself are unique compared with related structures inother capillary beds. The proximal component layerof the glomerular filter itself is a fenestrated endo-thelium, characterized by the presence of individualfenestrae on the order of 70–100 nm in diameter.These cells drape the luminal aspect of the capillaryand permit filtration. The second layer of the filter,the glomerular basement membrane (GBM), is a com-plex mesh of extracellular proteins, including type IVcollagen, laminins, fibronectins, and proteoglycans.The distal layer of the glomerular filter is composedof the visceral epithelial cells, or podocytes. These re-markable cells help to create the filtration slit dia-phragm and serve as support to help sustain theintegrity of the free-standing capillary loops. A thirdcell type, the mesangial cells, also contributes to theintegrity of the glomerular tuft and the dynamic na-ture of filtration. Together, this elegant structure per-mits the formation of the primary glomerular filtratethat enters a space delimited by the visceral and pa-rietal epithelial cells before modification during tran-sit through the tubule.

Hemodynamic Control of Glomerular Filtrationat the Single Nephron Level

Although there is considerable variability, the averagehuman kidney is composed of approximately 1 millionindividual functioning nephrons, each containing asingle glomerulus or filtering unit (1). The GFR for the

entire organism is then the sum of the individual fil-tration rates of approximately 2 million glomeruli intwo kidneys. There can be significant differences inthe size and filtration rates of individual glomeruliin different regions of the kidney. For example, juxta-medullary nephrons tend to have a larger intraglo-merular volume compared with superficial nephrons(2). Intrarenal blood flow distribution varies in phys-iologic and pathologic situations and this can also af-fect total GFR for the organism (3,4). Indeed, medullaryblood flow appears to be greatest during a water di-uresis and reduced in the setting of antidiuresis (4,5),and also appears to be better preserved in renal hypo-perfusion. This discussion focuses on the regulation offiltration at the level of a single, typical glomerulus, bywhich the precise hemodynamic control of glomerularfiltration is best understood. Although the majority ofthe experimental findings were gleaned from workdone in Wistar rats (which have superficial glomeruli,in contrast with most mammals), the principles re-viewed here are thought to apply to human glomerulias well.

The Determinants of Glomerular UltrafiltrationThe transcapillary movement of water across the

glomerular capillaries is controlled by the same Starlingforces that control fluid movement across all capillarybeds. Filtration occurs because there is an imbalancebetween the mean transcapillary hydraulic pressuregradient (D�P), which favors filtration, and the meantranscapillary oncotic pressure (D�p), which opposes fil-tration. (Of note, we use the term hydraulic, rather thanhydrostatic, because the latter refers to pressure in astatic liquid, whereas hydraulic pressure refers to pres-sure of fluid moving through tubes.) The resulting netpressure gradient integrated over the entire filteringportion of a single glomerular capillary networkis the mean net ultrafiltration pressure (�PUF). Thisphysiologic condition can be expressed mathemati-cally by Equation 1:

*Beth IsraelDeaconess MedicalCenter, Boston,Massachusetts;†Feinberg School ofMedicine,NorthwesternUniversity, Chicago,Illinois; and ‡BrownMedical School,Brown University,Providence, RhodeIsland

Correspondence:Dr. Martin Pollak, BethIsrael DeaconessMedical Center, WestCampus/Farr 8, 185Pilgrim Road, Boston,MA 02215. Email:mpollak@bidmc.harvard.edu

www.cjasn.org Vol 9 August, 2014 Copyright © 2014 by the American Society of Nephrology 1461

SNGFR 5 Kf 3��PUF

�

5 Kf 3�D�P2D�p

�

5 Kf

���PGC 2PT

�2 ð�pGC 2pT Þ

�

5 k 3 S3���PGC 2PT

�2 ð�pGC 2pT Þ

�(1)

where SNGFR is the single nephron GFR, �PGC is the hy-draulic pressure within the lumen of the glomerular capil-lary that is generated by the pumping action of the heart,and PT is the hydraulic pressure in Bowman’s space that istypically measured as the pressure within an early proxi-mal tubule segment. �pGC and pT are the oncotic pressureswithin the glomerular capillary, generated by the plasmaproteins, and in Bowman’s space. Because the sievingcharacteristics of the glomerular capillary wall that preventtranscapillary filtration of all but small, low molecularweight plasma proteins, pT is effectively zero. Kf is de-termined by two factors: the hydraulic conductivity of theglomerular capillary wall, which depends on both its in-trinsic characteristics and the concentration polarization ofproteins (6) in the capillary (k), and the total capillarysurface area available for filtration (S). For any given meannet filtration pressure, �PUF, the absolute amount of filtrateformed will also depend on Kf.

Glomerular Transcapillary Hydraulic and OncoticPressures, D�P and D�p. These relationships can also berepresented graphically as shown in Figure 2. In this con-struct, the glomerular capillary network is imagined to be acontinuous single tube beginning at the termination of theafferent arteriole, and ending at the origin of the efferentarteriole. The x axis represents fractional distance along theglomerular capillary. Both hydraulic and plasma oncoticpressures are shown on the y axis. The values provided inFigure 2 are typical of those observed in rats, which werefirst directly measured by Brenner et al. (7) and subsequentlyconfirmed in many studies in rats (8) and squirrel monkeys(9). PGC averages just ,50 mmHg in normal rats and isnearly constant along the glomerular capillary network, avalue significantly higher than in systemic capillaries. Theoncotic pressure Dp rises from approximately 18 to 34mmHg by the efferent end of the network, a consequenceof the filtration process during which water leaves the glo-merular capillary lumen causing an increase in protein con-centration. pGC equals DP by the end of the capillary,resulting in a reduction of local filtration pressure from 17to 0 mmHg (Figure 2).The oncotic pressure profile is highly nonlinear and is a

consequence of (1) the greater local ultrafiltration pressurenear the afferent end of the capillary leading to more rapidtranslocation of fluid and (2) the exponential relationship

Figure 1. | Structure of the renal corpuscle, looking into the Bowman’s capsule at glomerular capillary tuft. The capsule is linedwith parietalepithelium, which gives way to the cells of the proximal tubule at the urinary pole on the right. At the left, the vascular pole of the glomerulusincludes both the afferent and efferent arterioles. In addition, the relationship between these arterioles and the specialized portion of the distalnephron called the macula densa is illustrated. (Inset) The layers that comprise the filtration barrier are displayed. The outermost layer iscomposed of the visceral epithelial cells, the podocytes, next the glomerular basementmembrane (GBM) and finally the fenestrated endothelialcells.

1462 Clinical Journal of the American Society of Nephrology

between the plasma protein concentration and the plasmaoncotic pressure (8). In fact, under equilibrium conditions,the exact Dp curve cannot be determined. Curve A in Figure 2is just one of an infinite number of possibilities.Ultrafiltration Coefficient, Kf. As shown in Equation 1,

SNGFR equals the Kf multiplied by the net ultrafiltration pres-sure integrated over the entire length of the glomerular capil-lary. The mean net ultrafiltration pressure is representedgraphically in Figure 2 as the area between the DP and theDp curves (shaded blue). As discussed above, because an ex-act Dp curve cannot be determined under the normal condi-tion of filtration pressure equilibrium, the exact values for �PUF

and Kf can also not be determined. However, a minimumvalue for Kf can be estimated using the dashed line in Figure2 curve B, which gives a maximal possible value for �PUF.Applying this approximation to values for SNGFR and D�Pobtained in rats, Kf has been estimated to range from at least3.5 to 8 nl/min per mmHg. Using the value of S obtained forrat glomeruli, the calculated value of k is approximately 2500nl/min per mmHg per cm2. This value is one to two orders ofmagnitude greater than reported for other capillary beds (8),and this allows for the high rate of glomerular filtrationdespite a net driving pressure of,10 mmHg, on average, alongthe capillary.

The Effects of Selective Alterations in the PrimaryDeterminants of SNGFRThe four principal determinants of glomerular filtration at

the single nephron level are the Kf, the transcapillary

hydraulic pressure difference (D�P), the initial capillary on-cotic pressure (pA), and the initial glomerular plasma flowrate (QA) (Figure 3). Selective alteration in any of these fourprimary determinants has predictable effects on SNGFR,has been examined by mathematical modeling (10) and ex-perimentally (8), and is described here. However, physio-logic and pathophysiologic states often engender complexcombinations of alterations in multiple determinants thatmay be additive or offsetting. Indeed, these determinantsare not truly independent variables; rather, they tend tohave complex and often reciprocal relationships.Glomerular Capillary Ultrafiltration Coefficient, Kf.

Under conditions of filtration pressure equilibrium, whichare typically present when QA is normal to low, reductions inKf will not affect SNGFR (Figure 3A). Rather, because lessfluid exits the capillary at each point along the network, theoncotic pressure curve is shifted down and to the right, mov-ing the equilibrium point closer to the end of the capillary, asif going from curve A to curve B in Figure 2. This change inthe oncotic pressure profile results in an increase in the areabetween the DP and Dp, which is equivalent to an increasein the mean net ultrafiltration pressure for the entire capil-lary network (�PUF). This increase in �PUF exactly opposes andoffsets the decrease in Kf, resulting in constancy in SNGFR.Once Kf is reduced enough to produce disequilibrium, typ-ically by $50%, further reductions in Kf will be associatedwith declines in SNGFR.Because Kf is the product of the surface area available

for filtration and hydraulic conductivity, reductions in ei-ther can affect Kf. Reductions in glomerular surface areamay occur in glomerular diseases, such as membranousnephropathy (11) or physiologically, as a result of vaso-constrictors, such as NE (12). Hydraulic conductivity mayalso decline in glomerular diseases. This has been wellcharacterized in a model for membranous nephropathy(13) and in anti-GBM–induced nephritis (14). In addition,studies in patients with diabetic nephropathy suggest thatthe hydraulic conductivity is progressively decreased (15).Furthermore, changes in D�P (16) or a change in the plasmaprotein concentration (16) (see below) have also beenshown to alter hydraulic conductivity.Transcapillary Hydraulic PressureDifference,D�P. Until

DP exceeds pA, no filtration will occur (Figure 3B). Once thehydraulic pressure exceeds the oncotic pressure, filtration riseswith increases in D�P. This relationship is not linear becauseelevations in D�P are associated with more fluid filtration ear-lier in the capillary as well as a more rapid rise in Dp, al-though not enough to entirely offset the increase in D�P.However, because of the offsetting increase in oncotic pressureand because alterations in D�P of approximately .10 mmHgare unusual under physiologic conditions (10), changes in D�Pgenerally result in relatively minor changes in SNGFR.Oncotic Pressure, pA. SNGFR is predicted to vary inver-

sely with selective changes in pA (10) (Figure 3C). This isbecause increases in oncotic pressure should reduce �PUF

and, thereby, SNGFR. However, changes in pA do not appearto be an important factor in physiologic regulation of glo-merular filtration. First, the plasma protein concentration isrelatively stable except in disease states that alter either theproduction (e.g., in severe liver disease) or degradation or lossof plasma proteins (e.g., in nephrotic syndrome). In addition,changes in plasma protein concentration in mammals are

Figure 2. | Hydraulic and oncotic pressures along an idealized glo-merular capillary and within Bowman’s space. The graph shows ideal-ized hydraulic and oncotic pressure curves under conditions of filtrationpressure equilibrium (curve A) when the mean net ultrafiltration pressure(�PUF) is equal to the shaded region between the DP and the Dp curves.Curve B is a hypothetical linear profile of the oncotic pressure that wouldoccur with a minimum value of the Kf. By contrast, in curve C, filtrationequilibrium is never reached. Modified from Taal MW, Chertow GM,Marsden PA, Skorecki K, Yu ASL, Brenner BM: Brenner and Rector’s TheKidney, 9th Ed., Philadelphia, Elsevier Saunders, 2012, with permission.

Clin J Am Soc Nephrol 9: 1461–1469, August, 2014 Control of Glomerular Filtration, Pollak et al. 1463

associated with reciprocal and offsetting alterations in Kf, suchthat GFR is often unchanged. This is thought to be a result ofvariation in the hydraulic conductivity of the GBM in responseto changes in serum albumin concentration (16), but the exactmechanism is unknown. By contrast, GFR may be reduced inpathologic states in which the oncotic pressure is very high andexceeds hydraulic driving pressure. Case reports of AKI fromdextran infusions (17) and massive albumin infusions (18) sup-port this possibility.Glomerular Plasma Flow Rate, QA. Because glomerular

filtrate is essentially free of larger plasma proteins, conservationof mass requires that the total amount of protein entering thecapillary is equal to the amount leaving it (Figure 3D), which isexpressed mathematically in Equation 2:

QACA5QECE (2)

where QA and QE are the afferent and efferent arteriolarplasma flow rates and CA and CE are the respective afferentand efferent arteriolar blood plasma protein concentrations,respectively. Because QE5(QA–SNGFR), Equation 2 can berewritten as follows (Equation 3):

QACA5ðQA 2 SNGFRÞCE; (3)

which can be rearranged as follows (Equation 4):

SNGFR5ð12 ðCA=CEÞÞQA (4)

Although the relationship between the plasma proteinconcentration C and the oncotic pressure �p is not linear, Equa-tion 4 is approximately equal to the following (Equation 5):

SNGFR5ð12 ðpA=pEÞÞQA (5).

Because under conditions of filtration pressure equilibrium,pE5D�P as follows (Equation 6):

SNGFR5�12

�pA

�D�P

��QA (6).

Therefore, if pA and D�P are held constant, SNGFR will varydirectly with QA.This relationship can also be understood graphically by

examining Figure 2. As the plasma flow rate increases,there will be less contact time between the plasma flowingthrough the capillary and any given point along the cap-illary network. In that case, for any given pressure gradi-ent at any given point, less fluid will leave the capillary asfiltrate. As a result, the oncotic pressure profile will be shifteddown and to the right, as if moving from curve A to curve B.The equilibrium point is shifted closer to the efferent end ofthe capillary, resulting in an increase in �PUF and SNGFR. Onthe other hand, once QA is sufficiently large, disequilibriumwill occur, leading to less and less dependence of SNGFR onchanges in QA. In fact, for animals or under conditions thatare far from equilibrium, changes in QA are not predicted tohave major effects on SNGFR.

Afferent and Efferent Arteriolar Resistances, RA and RE

In most physiologic settings, alterations in pA, Kf, and PT

contribute little to the minute-to-minute regulation ofSNGFR. Instead, the filtration rate is regulated by changesin �PGC and QA, and they in turn are controlled by alterationsin afferent and efferent arteriolar resistances (RA and RE,respectively). The glomerular capillaries are unique in thatthey are arranged in series between two resistance vessels,the afferent and efferent arterioles. Selective modulation ofthe relative resistances in these two vessels allows for precise

Figure 3. | The single nephron GFR (SNGFR) can be affected by alterations in each of the four major determinants of ultrafiltration. Theseaffects can be expressed in amathematical model, which helps illustrate the predicted effects of each change on the SNGFR. Normal values areindicatedwith a dotted line in each graph.Unless otherwise indicated in the graph, the input values areKf5 0.095 nl/szmmHg,D�P5 35mmHg,pA

5 18mmHg, andQA5 135 nl/min. (A) Alterations in Kf; (B) alterations inDP; (C) alterations inpA; (D) alterations inQA.Modified fromBrennerBM,Dworkin LD, Ichikawa I: Glomerular filtration. In: Brenner and Rector’s The Kidney, 3rd Ed., edited by Brenner BM, Rector FC Jr, Philadelphia,Elsevier, 1986, with permission.

1464 Clinical Journal of the American Society of Nephrology

and largely independent regulation of �PGC and D�P, QA, andSNGFR. In addition, a large number of hormones, vasoac-tive substances, growth factors, cytokines, and drugs alterresistance in these vessels, selectively or in concert (12).

Renal AutoregulationOne important characteristic of the glomerular circula-

tion is autoregulation, by which not only QA but also D�Pand SNGFR are held constant over a wide range of renalartery pressures (Figure 4). Robertson et al. examined theprecise glomerular hemodynamic mechanisms that ac-count for renal autoregulation (10) by applying micro-puncture techniques to rats. Graded reductions in renalartery pressure were first associated with large declinesin RA, followed by significant increases in RE. By contrast,an increase in renal plasma flow from vasodilatory sub-stances was associated with a decline in RE and an increasein QA but was offset by a decrease in Kf. These combina-tions are predicted to be associated with near constancy ofD�P, QA, and SNGFR, which were indeed observed by theseauthors for renal artery pressures from approximately 80to 120 mmHg.

The mediators responsible for renal autoregulation in-clude a myogenic response that is intrinsic to the bloodvessels as well as tubuloglomerular feedback by whichchloride uptake by the macula densa segment of the distaltubule is sensed and then elicits effector pathways thatmodulate RA and RE and maintain the SNGFR (3). In ad-dition, changes in the filtration fraction will affect the on-cotic pressure in peritubular capillaries and contribute toglomerular tubular balance (19).

Glomerular Filtration of MacromoleculesClassic studies on the separation of the urinary filtrate from

the blood highlighted the restriction of serum proteins fromthe Bowman’s space. Detailed in vivo studies in both animalsand humans have been able to characterize the sieving prop-erties of the glomerular filter, which confers both size andcharge selectivity. Studies using dextran molecules of vary-ing sizes and charge demonstrated that neutral particleswith a molecular radius .4.2 nm are restricted from theurinary space, whereas anion particles .3.4 nm have a frac-tional clearance that approaches zero. Thus, the charge ofalbumin, which is approximately 3.6 nm in size, explainsthe limited clearance observed (20). By contrast, in both ex-perimental animals and humans with nephrotic syndrome,charge selectivity is lost with the alterations in normal glo-merular architecture and this appears to contribute to theprotein losses seen in these conditions (21).

Cellular and Molecular Components of theGlomerular FilterRecent years have seen a rapid increase in knowledge

about how the glomerulus and glomerular filtration areregulated at the cellular and molecular levels (Figure 5).Studies of diseases of the three basic components of theglomerular filter have provided a better understanding ofthe molecular physiology and pathophysiology of thisstructure. Much of this progress has come through studiesof inherited disorders of glomerular function. The glomer-ular filter is both a charge and size barrier. Movement ofanionic molecules across the filter is relatively restricted.Movement of large molecules across the filter is also re-stricted. The importance of all three components to normalcharge and size selectivity is clear from the fact that dis-ruption of any of these three components leads to proteinuricdisease states (22).

Disorders of Slit Diaphragm and Podocyte Structure andFunctionThe glomerular epithelial cell, or podocyte, has been the

focus of a great deal of recent investigation. These cells,which are derived from metanephric mesenchyme, formthe distal component of the glomerular filter and arebothmorphologically and functionally unique. Long intricateextensions, or primary processes, lead to secondary orfoot processes, which form a complex interdigitating struc-ture with the foot processes from adjacent podocytes. Theserpentine cell–cell junction known as the slit diaphragmthat bridges these podocytes can be considered a modifiedadherens junction with some unique additional molecularcomponents (23).

Figure 4. | The effects of graded reduction in renal artery perfusionpressure on glomerular hemodynamics in the rat. Glomerular bloodflow (GBF) and glomerular capillary pressure (�PGC) remained rela-tively constant over the range of pressure from 80 to 120 mmHg, inresponse to a marked decrease in afferent arteriolar resistance (RA).With further reduction in perfusion pressure to 60 mmHg, GBF de-clined proportionally more than �PGC primarily because of an increasein RE. Reprinted from Taal MW, Chertow GM, Marsden PA, SkoreckiK, Yu ASL, Brenner BM: Brenner and Rector’s The Kidney, 9th Ed.,Philadelphia, Elsevier Saunders, 2012, with permission.

Clin J Am Soc Nephrol 9: 1461–1469, August, 2014 Control of Glomerular Filtration, Pollak et al. 1465

Alterations in the structure and/or function of thevisceral glomerular epithelial cell, or podocyte, lead to aspectrum of human disease. The best established examplesare inherited disorders caused by changes in single genes.These inherited “podocytopathies” display a range of phe-notypes. As a general rule, recessive forms of podocytedysfunction are caused by mutations resulting in loss ofthe function of a normal protein.The proteins nephrin and podocin are essentially podocyte

limited in their expression. Humans, as well as mice, that lacknormal copies of either protein develop severe glomerulardisease. Both of these molecules were unknown until theywere identified on the basis of human genetic studies. Nephrin,a transmembrane domain protein with a large extracellulardomain, is required for both normal slit diaphragm structureand function. In Finland, two specific founder mutations inthe NPHS1 nephrin gene make this form of congenital ne-phrotic syndrome much more common than in other parts ofthe world (24,25). In general, nephrin-mediated congenitalnephrotic syndrome manifests as massive proteinuria ($20 g)in the neonatal period. Defects in the NPHS2 podocin genecause a wider spectrum of disease, ranging from congenitalnephrosis to adult-onset FSGS (25,26). Podocin, like nephrin,is an integral membrane protein. Podocin interacts directlywith nephrin and appears to help regulate its trafficking tothe slit diaphragm (27).The function of the podocyte in vivo, and consequently the

function of the entire glomerulus, is very sensitive to mildperturbations in its actin cytoskeleton (28). Mutations in thewidely expressed actin-regulating proteins a-actinin-4 andINF2 lead to human forms of podocyte dysfunction, FSGS,and progressive kidney disease, despite having little or noeffect on other organ function (29,30). Mutations in theTRPC6 cation channel cause a phenotypically similar auto-somal dominant form of FSGS (31).A good deal of recent work has focused on signaling

pathways critical to podocyte function. For example, fineregulation of small GTPase signaling appears essential formaintenance and repair of normal podocyte structure andfunction (32,33). Autophagy, a catabolic process of cell

“self-eating,” whereby cells degrade unnecessary or dys-functional intracellular components, is necessary for nor-mal podocyte function. Transgenic mice deficient inautophagy show marked glomerular pathology (34). Thecommunication between slit diaphragm proteins and thecytoskeleton is similarly of critical importance. Properfunctioning of the cytoskeletal apparatus is required fororganization of the slit diaphragm complex, and normalslit diaphragm function (as both a structural apparatusand signaling center) regulates actin dynamics (34,35).The functions of some glomerular disease genes and gene

products remain unclear. Variants in the APOL1 gene, encod-ing apoL1, explain the high rate of FSGS and other forms ofnondiabetic kidney disease in African Americans. Its functionin normal human physiology, if any, is unclear. APOL1 doesnot exist in rats, mice, and other nonprimates, and is present inonly a few primate species.

The GBMThe importance of the GBM to normal glomerular phys-

iology is made evident by the existence of several humandiseases characterized by abnormalities in the GBM struc-ture. The GBM is formed by the fusion of basement mem-branes from both the endothelial and epithelial cells thatform the glomerular corpuscle. Morphologically, the GBMcontains a dense inner layer called the lamina densa,flanked by the thinner laminae rara interna and rare externa.This morphology, as well as GBM function, is altered in thepresence of mutations in type IV collagens, the major com-ponents of the GBM. Recessive forms of Alport syndrome canbe caused by inheritance ofmutations in the autosomal type 4collagen genes COL3A4 and COL4A4, whereas the morecommon X-linked form is caused by mutations in COL5A4(36). Infants born with two mutant copies of the gene encod-ing the GBM protein laminin b2 can exhibit congenital ne-phrotic syndrome (37).Studies in severalmousemodels of alteredGBMcomponents

(type IV collagen, laminin) show that the altered glomerularfilters in these models can show increased permeability to largemolecules (38). These alterations are seen to occur before ultra-structural abnormalities in the podocyte (39). Thus, alteredGBM composition itself can lead to proteinuria. Recent studieshave also readdressed the issue of whether glycosaminogly-cans in the endothelium help regulate glomerular permeability,confirming the role of these large molecules in charge and sizeselectivity (40,41).Despite extensive study, there are still open questions

regarding the complex nature of this filter as well as how itallows such high rates of water flow while restricting theflow of large molecules and yet remains functional and“unclogged.” One recent hypothesis suggests that the glo-merular filter, and the GBM in particular, acts more like agel than a simple filter in that the size-selective propertiesof the glomerular filter are determined by permeation anddiffusion properties of the GBM (42).

EndotheliumThe glomerular endothelium represents the first layer in

the glomerular filtration barrier and is in direct contact withthe blood. Although the origin of these cells has been debatedfor some time, recent lineage tracing studies demonstrated

Figure 5. | Major molecular components of the podocyte and slit di-aphragm. Interdigitating podocytes from neighboring cells form theelaborate slit diaphragm that is composed of nephrin. Podocin helpsregulate trafficking of nephrin to the slit diaphragm. Proteins a-actinin-4and INF2play important roles in themaintenanceof theactin cytoskeleton,whereas integrins help anchor the podocytes to the glomerular basementmembrane.Mutations ina-actinin-4, INF2, and TRPC6 channel all causeautosomal dominant forms of FSGS. Mutations in nephrin and podocincause recessive forms of steroid-resistant nephrosis.

1466 Clinical Journal of the American Society of Nephrology

that they may derive from the Foxd1-positive stromal celllineage (43). Endothelial cells migrate from the metanephricmesenchyme into the developing glomerular tuft at theS-shape stage of glomerulogenesis. Glomerular endothelialcells are attracted by vascular growth factors, such as vas-cular endothelial growth factor (VEGF)-A, that are producedby adjacent podocytes (44–46). As glomerular developmentproceeds, the endothelial cells become progressively flattenedand develop fenestrations (two features critical for mainte-nance of glomerular function), permitting high flux of waterand small solutes. The endothelium is covered by a glycoca-lyx that likely restricts the passage of large macromolecules.Alterations of the glycocalyx were reported in various mod-els of glomerular disease, including diabetic nephropathy(41). Studies in humans have been limited by incompleteknowledge of glycocalyx components and functions and bythe need for specialized fixation and staining methods forproper visualization.

Diseases of the Glomerular EndotheliumPrimary diseases of the glomerular endothelium may

result in rapid loss of kidney function. Thrombotic micro-angiopathies (TMAs) represent a heterogeneous group ofdisorders characterized by varying degrees of endothelialswelling known as endotheliosis, fibrin and platelet deposi-tion, splitting of the GBM, and red blood cell fragmentation(47,48). The molecular causes of a number of TMAs wererecently elucidated. Although thrombotic thrombocytopenicpurpura and hemolytic uremic syndrome (HUS) were onceconsidered to represent a spectrum of the same disease (49),deficiency of ADAMTS13 enzymatic function as a result ofcongenital mutations or antibody-mediated inhibition isnow known to be responsible for thrombotic thrombocyto-penic purpura (50). By contrast, HUSs are associated withdysregulation of the alternate pathway of complement. Mul-tiple mutations in genes that encode proteins required forthe regulation of complement activation cause atypical HUS(48). Activation of complement leads to perturbation of oth-erwise thromboresistant renal endothelial cells with resul-tant local damage and an influx of inflammatory cells.Although the most common form of HUS—diarrheal HUS,caused by Shiga toxin—has not been clearly linked to com-plement dysregulation, several small series have demon-strated benefit from complement inhibition in patientsduring a recent large outbreak of diarrheal HUS in Germany(51,52). These data raise the intriguing possibility that com-plement dysregulation is responsible for renal injury in com-mon forms of HUS as well. Thrombotic injury to theglomerular microvasculature is also a universal feature ofthe renal disease of preeclampsia and in subsets of patientstreated with anti-VEGF agents. Importantly, podocytes pro-duce large amounts of VEGF during development and infiltering adult glomeruli in order to nurture the endothelialcells despite the tide of glomerular filtrate. This paracrinefactor appears to be important for maintenance of the nor-mal endothelial cell structure on the basis of observationsthat inhibition of VEGF signaling within the glomerular en-dothelium may result in thrombotic injury. Human, mouse,and cell-based studies support this model (53,54). Finally,mutations in the D7E lipid enzyme were identified in fam-ilies presenting with atypical HUS and/or FSGS (55,56). En-dothelial and podocyte injury were observed in biopsies

from these patients, suggesting that phosphorylation of lipidsubstrates is crucial for glomerular health.In addition to TMAs, activation of the glomerular endo-

thelium can result in pathologic angiogenesis, enhancedpermeability and proteinuria, and upregulation of cell adhe-sion molecules involved in recruitment of inflammatorycells. These processes are documented in a large number ofcommon glomerular diseases, including diabetic nephropa-thy and vasculitides. Future studies are needed to determinethe molecular basis of endothelial dysfunction, which shouldprovide interesting new therapeutic targets for glomerulardisease.

MesangiumThemesangium refers to themesangial cells and thematrix

they produce. During glomerular development, mesangialcells migrate into the developing tuft under the influence ofPDGFb produced by the glomerular endothelial cells actingon the PDGFb receptor expressed by mesangial cells (57).Traditionally, mesangial cells were referred to as pericytesbecause they provide support to the adjacent capillary loops.However, podocytes also provide essential growth factorsand support to the endothelium and they also function asspecialized perivascular cells, making the glomerular micro-vasculature somewhat unique because it has intimate as-sociations with two different supporting cell types. Themesangial processes are filled with bundles of actin andmyosin-based microfilaments that extend to contact theGBM, in which they bind laminin a5 via integrin a3b1 andthe basal cell adhesion molecule (58). These processes provideprotection against glomerular pressure and may regulateglomerular capillary flow via contractile properties (59).The mesangial matrix produced by mesangial cells is

composed of a diverse group of proteins, including typeIII–VI collagens, heparin sulfate proteoglycans, and elasticfiber proteins including fibronectin, laminin, entactin, andfibrillin-1. Accumulation of the mesangial matrix andthickening of the GBM are features commonly observedin a number of glomerular diseases, including diabetic ne-phropathy. Under these settings, the matrix may includenormal as well as new components such as collagen I. Pro-liferative glomerulonephritides may include mesangial cellproliferation and mesangial deposits.

Newer MethodologiesMuch of the “classical” physiology of the mammalian glo-

merulus has been determined through studies in rats, par-ticularly the extensive studies that have defined the controlof filtration at the single nephron level. New experimentaltools are expanding the ability to examine glomerular pro-cesses. These tools include new animal models, more sophis-ticated cellular models, and better methods to visualizeglomerular function in existing rodent models. For example,multiphoton fluorescence microscopy enables examinationof glomerular function in vivo, and is now being applied inanalyses of models of disease (60). Such studies have con-firmed that only a small amount of large proteins is filteredby the glomerulus.Genetic manipulation of other model organisms, includ-

ing mice and zebrafish, permits in vivo assessment of genefunction. Using morpholino technology, it is possible to

Clin J Am Soc Nephrol 9: 1461–1469, August, 2014 Control of Glomerular Filtration, Pollak et al. 1467

rapidly generate “gene knockdowns” in fish and analyze ef-fects on the pronephros, which contains a single glomerulus.More precise modulation of the genome is possible with themultitude of cell and temporally controlled gene knockoutstrategies available in mice, and development of newer tech-nologies, such as phenotype-driven N-ethyl-N-nitrosourea(ENU) screens, Clustered Regularly Interspaced Short Palin-dromic Repeats technology, and short hairpin RNA(shRNA) knockdowns. ENU screening is a technique inwhich the ENU alkylating is used to induce mutations andthen the phenotype of the progeny can be studied and can-didate genes from phenotypes of interest can be identified.Clustered Regularly Interspaced Short Palindromic Repeatstechnology can be used for gene “editing” to assist in thestudy of specific genetic targets. Finally, shRNA knockdownis a technique that utilizes shRNA sequences to silence atarget gene of interest. Combining cell-based studies, modelsystems, and advances in “omics”will translate into tremen-dous growth in our knowledge about the physiology andregulation of the glomerulus.

DisclosuresNone.

References1. Hinchliffe SA, Sargent PH, Howard CV, Chan YF, van Velzen D:

Human intrauterine renal growth expressed in absolute numberof glomeruli assessed by the disector method and Cavalieriprinciple. Lab Invest 64: 777–784, 1991

2. Horster M, Thurau K: Micropuncture studies on the filtration rateof single superficial and juxtamedullary glomeruli in the ratkidney. Pflugers Arch Gesamte Physiol Menschen Tiere 301:162–181, 1968

3. Gong R, Dworkin LD, Brenner BM, Maddox DA: The renal cir-culations and glomerular ultrafiltration. In: Brenner & Rector’sThe Kidney, edited by Brenner BM, Philadelphia, ElsevierSaunders, 2008, pp 91–129

4. Stein JH, Boonjarern S, Wilson CB, Ferris TF: Alterations in intra-renal blood flow distribution. Methods of measurement and re-lationship to sodium balance. Circ Res 32[Suppl 1]: 61–72, 1973

5. Blantz RC, Wallin JD, Rector FC Jr, Seldin DW: Effect of variationin dietary NaCl intake on the intrarenal distribution of plasmaflow in the rat. J Clin Invest 51: 2790–2795, 1972

6. Deen WM, Robertson CR, Brenner BM: Concentration polari-zation in an ultrafiltering capillary. Biophys J 14: 412–431, 1974

7. Brenner BM, Troy JL, Daugharty TM: The dynamics of glomerularultrafiltration in the rat. J Clin Invest 50: 1776–1780, 1971

8. Maddox DA, Deen WM, Brenner BM: Glomerular Filtration. In:Handbook of Physiology Renal Physiology, edited byWindhagerEE, Giebisch G, Baltimore, Williams and Wilkins, 1992

9. Maddox DA, Deen WM, Brenner BM: Dynamics of glomerularultrafiltration. VI. Studies in the primate. Kidney Int 5: 271–278,1974

10. Deen WM, Robertson CR, Brenner BM: A model of glomerularultrafiltration in the rat. Am J Physiol 223: 1178–1183, 1972

11. Squarer A, Lemley KV, Ambalavanan S, Kristal B, Deen WM,Sibley R, Anderson L, Myers BD: Mechanisms of progressiveglomerular injury in membranous nephropathy. J Am SocNephrol 9: 1389–1398, 1998

12. Dworkin LD, Ichikawa I, Brenner BM: Hormonal modulation ofglomerular function. Am J Physiol 244: F95–F104, 1983

13. Hladunewich MA, Lemley KV, Blouch KL, Myers BD: Determi-nants of GFR depression in early membranous nephropathy.Am JPhysiol Renal Physiol 284: F1014–F1022, 2003

14. Blantz RC, Wilson CB: Acute effects of antiglomerular basementmembrane antibody on the process of glomerular filtration in therat. J Clin Invest 58: 899–911, 1976

15. Tomlanovich S, Deen WM, Jones HW 3rd, Schwartz HC, MyersBD: Functional nature of glomerular injury in progressive di-abetic glomerulopathy. Diabetes 36: 556–565, 1987

16. Daniels BS, Hauser EB, Deen WM, Hostetter TH: Glomerularbasement membrane: In vitro studies of water and protein per-meability. Am J Physiol 262: F919–F926, 1992

17. MoranM, Kapsner C: Acute renal failure associatedwith elevatedplasma oncotic pressure. N Engl J Med 317: 150–153, 1987

18. Rozich JD, Paul RV: Acute renal failure precipitated by elevatedcolloid osmotic pressure. Am J Med 87: 359–360, 1989

19. Thomson SC, Blantz RC: Glomerulotubular balance, tubuloglo-merular feedback, and salt homeostasis. J Am Soc Nephrol 19:2272–2275, 2008

20. Brenner BM, Hostetter TH, Humes HD: Molecular basis of pro-teinuria of glomerular origin. N Engl J Med 298: 826–833, 1978

21. Guasch A, Deen WM, Myers BD: Charge selectivity of the glo-merular filtration barrier in healthy and nephrotic humans. J ClinInvest 92: 2274–2282, 1993

22. Haraldsson B, Nystrom J, DeenWM: Properties of the glomerularbarrier andmechanisms of proteinuria. Physiol Rev 88: 451–487,2008

23. Reiser J, Kriz W, Kretzler M, Mundel P: The glomerular slit di-aphragm is a modified adherens junction. J Am Soc Nephrol 11:1–8, 2000

24. Kestila M, Lenkkeri U, Mannikko M, Lamerdin J, McCready P,Putaala H, Ruotsalainen V, Morita T, Nissinen M, Herva R,Kashtan CE, Peltonen L, Holmberg C, Olsen A, Tryggvason K:Positionally clonedgene fora novel glomerular protein—nephrin—is mutated in congenital nephrotic syndrome.Mol Cell 1: 575–582, 1998

25. Tryggvason K, Patrakka J, Wartiovaara J: Hereditary proteinuriasyndromes and mechanisms of proteinuria. N Engl J Med 354:1387–1401, 2006

26. Boute N,Gribouval O, Roselli S, Benessy F, LeeH, Fuchshuber A,Dahan K, Gubler MC, Niaudet P, Antignac C: NPHS2, encodingthe glomerular protein podocin, is mutated in autosomal re-cessive steroid-resistant nephrotic syndrome. Nat Genet 24:349–354, 2000

27. Nishibori Y, Liu L, Hosoyamada M, Endou H, Kudo A, TakenakaH, Higashihara E, Bessho F, Takahashi S, Kershaw D,Ruotsalainen V, Tryggvason K, Khoshnoodi J, Yan K: Disease-causingmissense mutations in NPHS2 gene alter normal nephrintrafficking to the plasma membrane. Kidney Int 66: 1755–1765,2004

28. Faul C, Asanuma K, Yanagida-Asanuma E, Kim K, Mundel P:Actin up: Regulation of podocyte structure and function bycomponents of the actin cytoskeleton. Trends Cell Biol 17: 428–437, 2007

29. Kaplan JM, Kim SH, North KN, Rennke H, Correia LA, Tong HQ,Mathis BJ, Rodrıguez-Perez JC, Allen PG, Beggs AH, Pollak MR:Mutations in ACTN4, encoding alpha-actinin-4, cause familialfocal segmental glomerulosclerosis.Nat Genet 24: 251–256, 2000

30. Brown EJ, Schlondorff JS, Becker DJ, Tsukaguchi H, Tonna SJ,Uscinski AL, Higgs HN, Henderson JM, Pollak MR: Mutations inthe formin gene INF2 cause focal segmental glomerulosclerosis.Nat Genet 42: 72–76, 2010

31. Winn MP, Conlon PJ, Lynn KL, Farrington MK, Creazzo T,Hawkins AF, Daskalakis N, Kwan SY, Ebersviller S, Burchette JL,Pericak-Vance MA, Howell DN, Vance JM, Rosenberg PB: Amutation in the TRPC6 cation channel causes familial focalsegmental glomerulosclerosis. Science 308: 1801–1804, 2005

32. Greka A, Mundel P: Cell biology and pathology of podocytes.Annu Rev Physiol 74: 299–323, 2012

33. Mouawad F, Tsui H, Takano T: Role of Rho-GTPases and theirregulatory proteins in glomerular podocyte function. Can JPhysiol Pharmacol 91: 773–782, 2013

34. Huber TB, Benzing T: The slit diaphragm: A signaling platform toregulate podocyte function. Curr Opin Nephrol Hypertens 14:211–216, 2005

35. George B, Holzman LB: Signaling from the podocyte in-tercellular junction to the actin cytoskeleton. Semin Nephrol 32:307–318, 2012

36. Hudson BG, Tryggvason K, Sundaramoorthy M, Neilson EG:Alport’s syndrome, Goodpasture’s syndrome, and type IV col-lagen. N Engl J Med 348: 2543–2556, 2003

37. Zenker M, Aigner T, Wendler O, Tralau T, Muntefering H, FenskiR, Pitz S, Schumacher V, Royer-Pokora B, Wuhl E, Cochat P,Bouvier R, Kraus C, Mark K, Madlon H, Dotsch J, Rascher W,

1468 Clinical Journal of the American Society of Nephrology

Maruniak-Chudek I, Lennert T, Neumann LM, Reis A: Humanlaminin beta2 deficiency causes congenital nephrosis with me-sangial sclerosis and distinct eye abnormalities.HumMol Genet13: 2625–2632, 2004

38. Suh JH, Miner JH: The glomerular basement membrane as abarrier to albumin. Nat Rev Nephrol 9: 470–477, 2013

39. JaradG, Cunningham J, ShawAS,Miner JH: Proteinuria precedespodocyte abnormalities inLamb2-/- mice, implicating the glo-merular basement membrane as an albumin barrier. J Clin Invest116: 2272–2279, 2006

40. Friden V, Oveland E, Tenstad O, Ebefors K, Nystrom J, NilssonUA, Haraldsson B: The glomerular endothelial cell coat is es-sential for glomerular filtration. Kidney Int 79: 1322–1330, 2011

41. JeanssonM,HaraldssonB:Morphological and functional evidencefor an important role of the endothelial cell glycocalyx in the glo-merular barrier. Am J Physiol Renal Physiol 290: F111–F116, 2006

42. Smithies O: Why the kidney glomerulus does not clog: A gelpermeation/diffusion hypothesis of renal function. Proc NatlAcad Sci U S A 100: 4108–4113, 2003

43. Sims-Lucas S, Schaefer C, Bushnell D, Ho J, Logar A, ProchownikE, Gittes G, Bates CM: Endothelial progenitors exist within thekidney and lung mesenchyme. PLoS ONE 8: e65993, 2013

44. Eremina V, SoodM, Haigh J, Nagy A, Lajoie G, Ferrara N, GerberHP, Kikkawa Y, Miner JH, Quaggin SE: Glomerular-specific al-terations of VEGF-A expression lead to distinct congenital andacquired renal diseases. J Clin Invest 111: 707–716, 2003

45. Eremina V, Cui S, Gerber H, Ferrara N, Haigh J, Nagy A, Ema M,Rossant J, Jothy S, Miner JH, Quaggin SE: Vascular endothelialgrowth factor a signaling in the podocyte-endothelial compart-ment is required for mesangial cell migration and survival. J AmSoc Nephrol 17: 724–735, 2006

46. EreminaV, BaeldeHJ,Quaggin SE: Role of the VEGF—a signalingpathway in the glomerulus: Evidence for crosstalk betweencomponents of the glomerular filtration barrier.Nephron, Physiol106: 32–37, 2007

47. Coppo P, Veyradier A: Thrombotic microangiopathies: Towards apathophysiology-based classification. Cardiovasc Hematol Dis-ord Drug Targets 9: 36–50, 2009

48. Noris M, Remuzzi G: Atypical hemolytic-uremic syndrome.N Engl J Med 361: 1676–1687, 2009

49. Remuzzi G: HUS and TTP: Variable expression of a single entity.Kidney Int 32: 292–308, 1987

50. Levy GG,NicholsWC, Lian EC, Foroud T,McClintick JN,McGeeBM, Yang AY, Siemieniak DR, Stark KR, Gruppo R, Sarode R,Shurin SB, Chandrasekaran V, Stabler SP, Sabio H, Bouhassira EE,Upshaw JD Jr, Ginsburg D, Tsai HM: Mutations in a member ofthe ADAMTS gene family cause thrombotic thrombocytopenicpurpura. Nature 413: 488–494, 2001

51. Lapeyraque AL, Malina M, Fremeaux-Bacchi V, Boppel T,Kirschfink M, Oualha M, Proulx F, Clermont MJ, Le Deist F,Niaudet P, Schaefer F: Eculizumab in severe Shiga-toxin-associated HUS. N Engl J Med 364: 2561–2563, 2011

52. Delmas Y, Vendrely B, Clouzeau B, Bachir H, Bui HN, Lacraz A,Helou S, Bordes C, Reffet A, Llanas B, Skopinski S, Rolland P,Gruson D, Combe C: Outbreak of Escherichia coli O104:H4haemolytic uraemic syndrome in France: Outcome witheculizumab [published online ahead of print November 28,2013]. Nephrol Dial Transplant doi: 10.1093/ndt/gft470

53. Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S,Libermann TA, Morgan JP, Sellke FW, Stillman IE, Epstein FH,Sukhatme VP, Karumanchi SA: Excess placental soluble fms-liketyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunc-tion, hypertension, and proteinuria in preeclampsia. J Clin Invest111: 649–658, 2003

54. Eremina V, Jefferson JA, Kowalewska J, Hochster H, Haas M,Weisstuch J, Richardson C, Kopp JB, Kabir MG, Backx PH,Gerber HP, Ferrara N, Barisoni L, Alpers CE, Quaggin SE: VEGFinhibition and renal thrombotic microangiopathy. N Engl J Med358: 1129–1136, 2008

55. LemaireM, Fremeaux-Bacchi V, Schaefer F, ChoiM, TangWH, LeQuintrec M, Fakhouri F, Taque S, Nobili F, Martinez F, Ji W,Overton JD, Mane SM, Nurnberg G, Altmuller J, Thiele H, MorinD, Deschenes G, Baudouin V, Llanas B, Collard L, Majid MA,Simkova E, Nurnberg P, Rioux-Leclerc N, Moeckel GW, GublerMC, Hwa J, Loirat C, Lifton RP: Recessive mutations in DGKEcause atypical hemolytic-uremic syndrome.Nat Genet 45: 531–536, 2013

56. Ozaltin F, Li B, Rauhauser A, An SW, Soylemezoglu O, Gonul II,Taskiran EZ, Ibsirlioglu T, Korkmaz E, Bilginer Y, Duzova A,OzenS, Topaloglu R, Besbas N, Ashraf S, Du Y, Liang C, Chen P, Lu D,Vadnagara K, Arbuckle S, Lewis D, Wakeland B, Quigg RJ,Ransom RF, Wakeland EK, Topham MK, Bazan NG, Mohan C,Hildebrandt F, Bakkaloglu A, Huang CL, Attanasio M: DGKEvariants cause a glomerular microangiopathy that mimicsmembranoproliferative GN. J Am Soc Nephrol 24: 377–384,2013

57. Bjarnegard M, Enge M, Norlin J, Gustafsdottir S, Fredriksson S,Abramsson A, Takemoto M, Gustafsson E, Fassler R, Betsholtz C:Endothelium-specific ablation of PDGFB leads to pericyte lossand glomerular, cardiac and placental abnormalities. De-velopment 131: 1847–1857, 2004

58. Kerjaschki D,Ojha PP, SusaniM,Horvat R, Binder S, Hovorka A,Hillemanns P, Pytela R: A beta 1-integrin receptor for fibro-nectin in human kidney glomeruli. Am J Pathol 134: 481–489,1989

59. Ghayur MN, Krepinsky JC, Janssen LJ: Contractility of the renalglomerulus and mesangial cells: Lingering doubts and strategiesfor the future. Med Hypotheses Res 4: 1–9, 2008

60. Peti-Peterdi J, Sipos A: A high-powered view of the filtrationbarrier. J Am Soc Nephrol 21: 1835–1841, 2010

Published online ahead of print. Publication date available at www.cjasn.org.

Clin J Am Soc Nephrol 9: 1461–1469, August, 2014 Control of Glomerular Filtration, Pollak et al. 1469