Evidence for restriction of fluid and solute movement across the glomerular capillary wall by the...

1

Evidence for restriction of fluid and solute movement across the glomerular capillary

wall by the sub podocyte space

Andrew HJ Salmon1,2, Ildikó Toma3, Arnold Sipos3, Robert Muston1, Steven J Harper1,

David O Bates1, Christopher R Neal1, János Peti-Peterdi3

1Microvascular Research Laboratories, Bristol Heart Institute, Department of

Physiology, University of Bristol, Southwell Street, Bristol, UK. BS2 8EJ. 2Academic

Renal Unit, Clinical Science at North Bristol, University of Bristol, Paul O’Gorman

Lifeline Centre, Southmead Hospital, Westbury-on-Trym, Bristol, UK. BS10 5NB. 3

Departments of Physiology, Biophysics & Medicine, Zilkha Neurogenetic Institute, Keck

School of Medicine, University of Southern California, Los Angeles, CA 90089, USA.

Running title: Physiological role of the sub podocyte space

Corresponding authors: Dr J Peti-Peterdi or Dr DO Bates

Addresses for Correspondence:

Dr J Peti-Peterdi, ZNI 335, Zilkha Neurogenetic Institute, 1501 San Pablo St., Keck

School of Medicine, University of Southern California, Los Angeles, CA 90089, USA.

[email protected] Tel: (001) 323 442 4337 Fax: (001) 323 442

4466

Dr DO Bates, Microvascular Research Laboratories, Bristol Heart Institute, Department

of Physiology, Southwell Street, University of Bristol, Bristol, BS2 8EJ, UK.

[email protected] Tel: (0044) 117 928 9818 Fax: (0044) 117 928 8151

of 38Articles in PresS. Am J Physiol Renal Physiol (September 12, 2007). doi:10.1152/ajprenal.00187.2007

Copyright © 2007 by the American Physiological Society.

2

Abstract The glomerular filtration barrier (GFB) is generally considered to consist of three layers:

fenestrated glomerular endothelium, glomerular basement membrane, and filtration slits

between adjacent podocyte foot processes. Detailed anatomical examination of the

GFB has revealed a novel abluminal structure, the sub-podocyte space (SPS), identified

as the labyrinthine space between the underside of podocyte cell body/primary

processes and the foot processes. The SPS covers 50-65% of the filtration surface of

the GFB, indicating that SPS may influence glomerular permeability. We have examined

the contribution of the SPS to the permeability characteristics of the GFB using

multiphoton microscopy techniques in isolated perfused glomeruli and in the intact

kidney in vivo. SPS were identified using this technique, with comparable dimensions to

SPS examined with electron microscopy. The passage of the intermediate weight

molecule rhodamine-conjugated 10kDa-dextran, but not the low weight molecule Lucifer

yellow (≈450 Da), accumulated in SPS-covered regions of the GFB, as compared with

GFB regions not covered by SPS (“naked regions”). Net Lucifer Yellow flux (taken to

indicate fluid flux) through identifiable SPS regions was calculated to be 66 - 75% of that

occurring through naked regions. These observations indicate both ultrafiltration and

hydraulic resistance imparted by the SPS, demonstrating the potential physiological

contribution of the SPS to glomerular permeability.

Keywords: Glomerulus; Permeability; Podocyte; Multiphoton

of 38

3

Introduction

The glomerular filtration barrier (GFB) has long been considered to comprise three

major elements (36). The luminal aspect of the barrier is formed by a continuous layer

of thin endothelial cells. Approximately 20% (5) of the surface area of these cells is

accounted for by transcellular channels of 60-80 nm diameter, termed fenestrae. It is

generally held that these fenestrae do not possess the diaphragm that bridges the

aperture in many other fenestrated capillary beds (18), although novel oxygen-carrying

fixative techniques have challenged this view (13). The endothelium of the GFB

contains an endothelial surface layer, or glycocalyx, the thickness of which has been

estimated as between 150 and 400 nm (13, 26). The glomerular basement membrane

(GBM) forms the middle portion of the barrier, and is itself tri-layered (lamina rara

interna & externa, and lamina densa). The outermost aspect of the GFB comprises the

regular interdigitations of podocyte foot processes. The highly conserved 43nm interval

(7) between adjacent foot processes is characterised by a ladder-like structure of

closely interweaved nephrin molecules (25, 34) that link to a host of slit diaphragm

complex associated molecules (2).

These three major components interact to generate the overall permeability coefficients

of the GFB. Isolated defects in any one of the three layers of the barrier can result in

alterations in permeability e.g. digestion of glomerular endothelial glycocalyx with

lumen-restricted enzymes (14), selective deletion of the important GBM component

laminin β-1 (22), and loss of critical slit diaphragm components, including nephrin (16).

These defects can result in glomerular nephropathies. Defects in the GBM are

of 38

4

recognised in conditions such as Alport’s Syndrome, and thin basement membrane

nephropathy (30), whereas mutations in molecules at the slit diaphragm such as

nephrin result in congenital nephropathies, e.g. of the Finnish type (16). Endothelial

dysfunction, e.g. in diabetes or pre-eclampsia, also results in increased glomerular

permeability and proteinuria (31). Nevertheless, no layer in isolation is considered to

dominate either the hydraulic resistance or the permselective properties of the barrier

(5), and both its anatomical characteristics (5) and its molecular constituents (9, 28)

ensure interdependence of all layers of this barrier.

In 2005, the anatomical pathway dictating the movement of primary ultrafiltrate from the

abluminal aspect of the podocyte filtration slits to the proximal tubule was re-evaluated

(20). What had hitherto been described as Bowman’s space (12, 17) was shown to

comprise three distinct urinary spaces, based on the identification of focal narrowings

within “Bowman’s space”. The first focal narrowing was identified between highly

attenuated and tortuous urinary space regions beneath podocyte cell bodies or foot

processes and between long narrow channels coursing through the glomerular tuft. This

narrowing therefore separated these regions into distinct urinary spaces: the sub-

podocyte space (SPS) and inter-podocyte space (IPS) respectively, and the narrowing

itself was termed the “sub-podocyte space exit pore” (SEP). The second narrowing was

demonstrated between the IPS and the shell-like urinary space between the glomerular

tuft and Bowman’s capsule, thereby distinguishing the IPS from the “peripheral urinary

space” (PUS). The SPS was found to overly between 50 and 65% of the GFB,

irrespective of whether or not the GFB was on the outer aspect of the glomerular tuft, or

of 38

5

deep within it (20). In a subsequent anatomical study of SPS dimensions, in which

fixation conditions required for electron microscopy were approximated as closely as

possible to those found during glomerular perfusion in vivo (21), the mean SPS height

was a mere 340 nm (approximately the same width as the entire underlying GFB), and

the average distance from GFB to the nearest SPS exit pore was 6700 nm (see figure

1A). Mathematical modeling, using anatomical information from the aforementioned

study and based on a circular flow model, indicated a hydraulic resistance of the SPS of

approximately 3-4 times that of the GFB (21). These observations led to the suggestion

that the movement of fluid across the SPS might be substantially lower than that

occurring across regions of the GFB that were not covered by SPS (“naked” regions). In

addition, the prevalence of SPS coverage of the GFB would render this differential fluid

flux highly physiologically significant. We therefore set out to test the hypothesis that

molecular movement across SPS-covered GFB would differ from that occurring across

naked regions of the GFB. This study uses multiphoton fluorescence imaging of isolated

microperfused glomeruli and the intact kidney in vivo since this technology can section

optically through an entire glomerulus with high resolution and without causing damage

to the living tissue (15).

Materials and Methods

Isolated microperfused glomerulus preparation

Experiments were performed according to protocols approved by the Institutional

Animal Care and Use Committee of University of Southern California. Animals [female

of 38

6

New Zealand rabbits, ≈ 500g (Irish farms, Norco, CA); n=6] were housed in a

temperature-controlled room with a 12 hour light-dark cycle with water and standard

chow available ad libitum. Anesthetised rabbits were euthanized by decapitation, and

left nephrectomy performed immediately post-mortem. Transverse slices of renal tissue

(2 mm thick) were immediately placed into cooled (4°C) dissection media [Dulbecco’s

Modified Eagle’s Medium/F12 (D8900, Sigma, St Louis, USA); 14.28mM NaHCO3;

(bubbled with 95%O2/5%CO2 for approximately 30 minutes); 3 g.dl-1 FBS; pH adjusted

to 7.40±0.02 with 1M NaOH]. Arcuate artery preparations that retained separated

lobules of renal cortical tissue were further dissected (Carl Zeiss Stemi-2000CS

stereomicroscope), to reveal afferent arteriole - glomerulus preparations, from which

redundant tubular tissue was carefully removed, taking care not to stretch the afferent

arteriole. The afferent arteriole was then severed at its base with a 30g needle, and the

afferent arteriole–glomerulus preparation transferred to a thermo-regulated Lucite

chamber on the experimental microscope stage. Once secured, fresh mammalian

modified Krebs-Ringer solution within the perfusion bath [mM: NaCl, 115; KCl, 5;

Na2HPO4, 0.24; NaH2PO4, 0.96; NaHCO3, 25; MgSO4, 1.2; CaCl2, 2; D(+) glucose, 5.5;

100 µM L-arginine (Sigma), pH 7.40±0.02 (95%O2/5%CO2)] was continuously

exchanged, and the temperature gradually elevated to 36±2°C. The afferent arteriole

was cannulated with a series of mounted concentric pipettes (Vestavia Scientific,

Vestavia, AL) containing Krebs-Ringer solution (as above) supplemented with 10 mg.ml-

1 BSA, and successful glomerular capillary perfusion was marked by the longitudinal

movement of red cells within the glomerular capillaries, and their expulsion from the

of 38

7

efferent arteriole. Additional details of the afferent arteriole-attached glomerulus

microperfusion technique have been described earlier (23,24).

In vivo imaging of the intact kidney

Munich-Wistar-Fromter male rats (200 g, Harlan, Madison, WI, n=3) were

anesthetized with thiobutabarbital (Inactin, 130 mg/kg body wt). The left kidney was

prepared for multi-photon imaging as described before (15). Briefly, after assuring

adequate anesthesia, the trachea was cannulated to facilitate breathing. The left

femoral vein and artery were cannulated for dye infusion and blood pressure

measurements, respectively. Subsequently, a 10-15 mm dorsal incision was made

under sterile conditions and the kidney was exteriorized. The animal was placed on the

stage of an inverted microscope with the exposed kidney placed in a coverslip-bottomed

heated chamber bathed in normal saline and the kidney was visualized from below

using a HCX PL APO 63X/1.4NA oil CS objective (Leica). During all procedures and

imaging, core body temperature was maintained with a homeothermic table. All animal

protocols have been approved by the Institutional Animal Care and Use Committee at

the University of Southern California.

Imaging system

Images were obtained via a 63x oil-immersion objective attached to a Leica DM IRE2

inverted microscope (Leica Microsystems, Heidelberg, Germany) with a number of

internal, external and transmitted light detector photomultipliers, and collected in time

(xyt) or z-section (xyz) series with Leica LCS imaging software. Conventional one-

photon excitation was achieved with a Leica TCS SP2 AOBS MP confocal microscope

of 38

8

system, powered by a series of lasers: red (HeNe 633 nm / 10 mW); orange (HeNe 594

nm/2 mW); green (HeNe 543 nm/1.2 mW); and blue (Ar 458 nm/5 mW; 476 nm/5 mW;

488 nm/20 mW; 514 nm/20 mW). Multiphoton excitation was achieved with an infrared

(710–920 nm) combined photo-diode pump laser and mode-locked titanium:sapphire

laser (Mai-Tai, Spectra-Physics, CA, USA). Further details of the imaging system are

given by Peti-Peterdi (24).

Dyes & tracers

Dyes: Cell nuclei were stained with 3 µM Hoechst 33342 (Molecular Probes, Eugene,

OR). Cell cytoplasm staining was obtained with 25 µM Quinacrine (Sigma). Dyes were

added to the perfusate solution in combination, and applied to the lumen of the perfused

afferent arteriole and glomerular capillaries for 20 – 30 min. During this loading time, an

appropriate section of glomerular capillaries were selected for subsequent experiments.

In in vivo experiments, Hoechst 33342 was used (10 µl of a 10 mg/ml stock in iv. bolus)

to label cell nuclei and a 500kDa dextran-fluorescein conjugate (Molecular Probes,

neutral, 20 µl of a 10 mg/ml stock in iv. bolus) labelled the circulating plasma.

Tracers: 0.3 mg.ml-1 lucifer yellow (MW 457Da) or 40 µg.ml-1 3kDa-dextran–Alexa 488

and 40 µg.ml-1 10kDa-dextran–rhodamine (all from Molecular Probes) served as low

and intermediate molecular weight markers of solute movement across the GFB

respectively. These relatively low molecular weight markers were chosen since both

should move across the GFB with little (dextran) or no (lucifer yellow) selective retention

(32). Because the ionic charge of fluorophores affects glomerular filtration

characteristics, neutral conjugates of dextran were preferentially used. Although the

of 38

9

fluid-marker LY is an anionic compound, it is freely filtered due to its small size (0.45

kD). Following dye loading, tracers were applied individually to the isolated perfused

glomerulus preparations. xyt-series of the initiation, perfusion and washout of each

tracer (perfusion: n=6 for each tracer) was obtained in three glomeruli; in one

glomerulus, Lucifer yellow and 10kDa-Dextran–rhodamine were perfused sequentially.

In some experiments the 3kDa and 10kDa dextrans were perfused simultaneously.

Individual images in the xyt-series were separated by <0.8 s intervals.

Hoechst 33342 (emission between 400-460 nm), quinacrine (emission at 510 nm),

dextran-Alexa 488 and dextran-fluorescein (emission at 520 nm), Lucifer yellow

(emission at 540 nm), and dextran-rhodamine (emission at 590 nm) were all excited at

860 nm. The emitted, non-descanned fluorescent light was detected by two external

photomultipliers (green (including blue) and red channels) with the help of a

FITC/TRITC filter block (Leica).

Image analysis

Regions of interest

xyt-series were analysed with Leica LCS imaging software. Spatial regions of interest

were selected, and the position verified in all images comprising the series (figure 1B).

Spatial regions were selected only on the outer aspect of peripheral capillary loops,

where abluminal cell bodies were taken to represent podocytes. GFB regions were

distinguished by the absence (“naked” regions figure 1Bi) or presence (“covered”

regions, figure 1Bii) of peripheral abluminal cell (podocyte) coverage. Covered regions

were taken to represent GFB with overlying SPS, as indicated by electron microscopy

of 38

10

studies (20) and the results reported herein. Any minor shift in the location of the

capillary wall region of interest in the multiphoton image between perfusion periods was

accommodated by duplication of the analysis region when necessary, and ensuring that

measurements were taken from exactly the same portion of the GFB in all images

comprising the xyt-series. Two (paired) regions of interest were studied in all cases: one

comprising GFB with or without SPS coverage, and one within an adjacent glomerular

capillary lumen.

Fluorescence intensity per unit area [ If ] of the dyes [dye If: indicating cellular

structures] and tracers [tracer If: representing solute molecules] were determined in the

region of interest in every frame of the series. dye If was carefully monitored in all

frames of the xyt-series, and constant values were taken to indicate constancy of the

anatomy of the region of interest. Baseline tracer If values were subtracted from all

subsequent tracer If measurements to facilitate comparison between tracer If at any

given time in the capillary lumen and capillary wall. tracer If values were plotted against

time.

Analysis of solute flux through SPS-covered regions

Linear regression lines were applied to tracer If measurements in the time period

encompassed by the beginning and end of the stable values of tracer If achieved in both

the capillary lumen and adjacent capillary wall regions. The slopes of these lines

describe the change in tracer If per unit time [d(tracer If)/dt]. Any change in d(tracer If)/dt

in the capillary wall that is not reflected by a comparable change in d(tracer If)/dt in the

capillary lumen indicates ultrafiltration of the tracer by the capillary wall. d(tracer If)/dt

of 38

11

within the capillary lumen and capillary wall regions were compared with ANCOVA

(Prism, GraphPad Software, CA, USA).

d(tracer If)/dt in the capillary lumen region [d(tracer If)/dt]lumen was subtracted from

d(tracer If)/dt in the adjacent capillary wall region [d(tracer If)/dt]wall for each pair of

regions studied, in order to eliminate any potential apparent change in tracer If in the

wall region attributable to a change in tracer If in the perfusate. These subtracted paired

measurements for all SPS-covered GFB regions and all naked GFB regions were

compared with one-way ANOVA with Bonferroni’s correction.

Analysis of fluid flux rate through SPS-covered regions

The absence of a rise in lucifer yellow If measurements in either covered or naked GFB

regions (figure 4) indicates absence of ultrafiltration of lucifer yellow, as predicted by its

molecular weight. Lucifer yellow flux was therefore taken as a surrogate marker of

solvent flux. Solvent flux through naked regions was taken as the reference value.

Calculation of solvent flux through the SPS, relative to that occurring through naked

GFB regions, is described in the results section.

Results

Anatomy of urinary spaces

Anatomical measurements of SPS were obtained from xyz-series of quinacrine-loaded

isolated perfused glomeruli. SPS were identified as non-fluorescent regions bounded by

GFB on one side, and by cell body (or cellular processes extending from the cell body)

of abluminal cells (figure 1B). Only abluminal cells overlying the outer aspect of

peripheral capillary loops were considered, in order to avoid confusion between

of 38

12

podocytes and mesangial cells. In the first set of studies using 3 glomeruli (from 3

different animals), 5 SPS were unequivocally identified by this method. Mean±S.E.M.

SPS dimensions were 2.3±0.6 µm depth x 4.0±0.5 µm width x 0.8±0.1 µm height. SPS

heights are comparable to those obtained in the electron microscopy study of Neal et al.

(20) under similar perfusion conditions.

Isolated glomeruli were loaded with quinacrine and perfused with lucifer yellow or

fluorescently labeled dextrans of different sizes (10kDa and 3kDa). Fig. 2 demonstrates

the feasibility of detecting the dynamics of solute filtration with high speed scanning and

to simultaneously resolve the anatomy of naked and SPS-covered GFB regions with

multi-photon imaging. The microperfused solutes quickly appeared in the lumen of

glomerular capillary loops and then filtered into the Bowman’s capsule. Within 5-10 s,

fluorescence intensity of solutes reached a maximum and stabilized in the capillary

lumen and Bowman’s space. Filtration of solutes were clearly visible across both naked

and SPS-covered regions of the GFB (Fig. 2).

Solute movement through SPS

10kDa Dextran (larger solute)

Representative plots of dextran If vs time are given in figure 3. The fluorescence

intensity rises as the solute moves into the region, and after the filling artefact stays

constant in the lumen and in naked GFB regions (i.e. GFB not covered by SPS)

([d(dextran If)/dt]naked 0.01 units.s-1; [d(dextran If)/dt]lumen 0.04 units.s-1: p>0.65,

ANCOVA, figure 3a), indicating that perfusion concentration is constant, and flux has

reached a steady state. In contrast, dextran If rose during the period of stable tracer

perfusion in the SPS-covered GFB region (figure 3b), indicating accumulation of 10kDa-

of 38

13

dextran-rhodamine within the SPS ([d(dextran If)/dt]SPS 0.82 units.s-1; [d(dextran

If)/dt]lumen 0.13 units.s-1: p<0.0001, ANCOVA).

Lucifer yellow (small solute)

Representative plots of LY If vs time are given in figure 4. In contrast to the selective

increase in dextran If in SPS-covered GFB regions, LY If remained approximately stable

in both naked (figure 4a) and SPS-covered (figure 4b) GFB regions throughout the

tracer perfusion period. In these examples, there was no significant difference between

d(LY If)/dt in capillary lumen ([d(LY If)/dt]lumen) or capillary wall ([d(LY If)/dt]wall) regions,

for either SPS-covered ([d(LY If)/dt]SPS 0.28 units.s-1; [d(LY If)/dt]lumen 0.21 units.s-1:

p>0.45, ANCOVA) or naked ([d(LY If)/dt]naked 0.18 units.s-1; [d(LY If)/dt]lumen 0.09 units.s-

1: p>0.60, ANCOVA) GFB regions.

Changes in tracer If in the capillary wall region during the tracer perfusion period may be

attributable to selective retention of the tracer by the components of the capillary wall, or

to changes in the concentration of tracer that is perfused. In order to eliminate any

influence of the latter, [d(tracer If)/dt]lumen was subtracted from [d(tracer If)/dt]wall for each

pair of wall and adjacent lumen regions for each experimental replicate. Mean±S.E.M.

([d(tracer If)/dt]wall - [d(tracer If)/dt]lumen) values are given in table 1, and plotted in figure

5. Only ([d(dextran If)/dt]wall - [d(dextran If)/dt]lumen) for 10kDa-dextran-rhodamine in SPS-

covered GFB regions was significantly different from zero (p<0.0001, one-sample t-test

vs zero; p>0.25, p>0.55 and p>0.75 for all other tracers / regions), indicating that

10kDa-dextran-rhodamine, but not lucifer yellow accumulated only in capillary wall

regions containing SPS. ([d(tracer If)/dt]wall - [d(tracer If)/dt]lumen) for 10kDa-dextran-

of 38

14

rhodamine in SPS-covered regions was significantly different from the same parameter

for 10kDa-dextran-rhodamine in naked regions, and also significantly different from

lucifer yellow values in both SPS and naked capillary wall regions (p<0.001 for 10kDa-

dextran-rhodamine in SPS regions vs all other tracers / regions, one-way ANOVA with

Bonferroni’s correction).

Simultaneous perfusion of 10kDa and 3kDa dextrans

To exclude the possibility that differences in chemical characteristics of the larger and

small fluorescent tracers (dextran and LY) were responsible for the difference in solute

fluxes, additional experiments were performed using simultaneous perfusion of 10kDa

and 3kDa dextrans in equivalent amounts. Representative plots of the two different size

dextran’s If vs time are given in Fig. 6. As with LY, the fluorescence intensity of the small

3kDa dextran quickly rose as the solute moves into the region, and then stayed

constant both in the naked GFB regions (Fig. 6a) and SPS-covered regions (Fig. 6b). In

contrast, the 10 kDa dextran If rose during the period of stable tracer perfusion in the

SPS-covered GFB region (Fig. 6b), but not in the naked region (Fig. 6a), indicating

accumulation of 10kDa-dextran-rhodamine within the SPS. Consequently, the ratio of

10kDa/3kDa dextrans (Fig. 6c) continuously rose above 1 in the SPS-covered region

(dRatio)/dtSPS 31.4 mUnits.s-1), but stayed constant around 1 in the naked regions

(dRatio/dtnaked 2.1 mUnits.s-1: p<0.0001, ANCOVA, n=4). These studies confirm that the

filtration characteristics of the two different small solutes, LY and the 3kDa dextran were

similar.

of 38

15

In vivo imaging of the 10kDa dextran (larger solute)

To confirm the presence and functional significance of SPS in the living kidney, in vivo

multi-photon imaging of the intact kidney was performed in Munich-Wistar-Fromter rats

as described previously (15). Representative images of a glomerulus in the intact rat

kidney are shown in Fig. 7 taken at the indicated time points after the iv bolus injection

of rhodamine B-conjugated 10kDa dextran. Appearance of the 10kDa dextran in

glomerular capillaries and its temporary accumulation in SPS-covered regions of the

filtration barrier during dye washout was visible (Fig. 7, see inset, ii arrow). The peak

ratio of fluorescence intensities in the naked or SPS-covered GFB regions vs. that in the

Bowman’s space were 0.97 ± 0.001 and 1.33 ± 0.002, respectively (n=6 each),

indicating the accumulation of the 10kDa dextran in the SPS, but not in the naked

regions of the GFB.

Fluid flux through SPS

Of interest was the observation that the intensity at which a steady state was reached

with lucifer yellow appeared to be closer to the luminal level in naked capillary wall

regions (figure 4a) as compared with covered capillary wall regions (figure 4b). The

mean ratio of LY If in naked regions (nakIf) to that in adjacent luminal regions (nakIf) was

0.65±0.07. The comparable ratio in covered regions (covIf / lumIf) was 0.46±0.06.

Since lucifer yellow is freely filtered, LY If in the non-restrictive urinary space outside

naked capillary wall regions is equal to that in the capillary lumen. The contents of the

naked capillary wall regions are the GFB (height from multiphoton images: 1.02±0.04

µm) and this non-restrictive urinary space (height 0.25±0.12). The 65% reduction in

of 38

16

mean nakIf / lumIf must therefore be attributable to the fraction of the GFB from which dye

is excluded (podocyte foot processes, endothelial cytoplasm and unavailable regions of

GBM: 44% of total GFB volume, see figure 4).

Covered GCW regions contain GFB, SPS and podocyte cell bodies/processes (total

mean height 1.55 µm). The observation that mean covIf / lumIf (0.46±0.06) is considerably

below mean nakIf / lumIf (0.65±0.07) may be attributable to complete dye exclusion from

the fraction of covered regions occupied by podocyte cell bodies/processes, or may be

due to reduced solvent flux through SPS (assuming that the ≈450 Da molecule lucifer

yellow is a surrogate marker for fluid.) Values for SPS height have been taken from the

literature (mean height under comparable perfusion conditions with low/no perfusate

oncotic pressure: 0.7 µm (20) ; mean height with physiological perfusate oncotic

pressure: 0.35 µm). The height of the cell body component of covered regions is taken

as the total covered region height less both GFB height and SPS height. Assuming that

44% of GFB volume excludes lucifer yellow (as for naked regions), the fraction of the

total volume of covered regions that is available to lucifer yellow is in the range 0.70

(SPS height 0.7 µm) to 0.62 (SPS height 0.35 µm). These fractions are considerably

higher than mean covIf / lumIf (0.46±0.06), indicating that there are fewer lucifer yellow

molecules per unit available volume in SPS than in the adjacent capillary lumen. This

indicates that solvent flux through SPS is in the range 66 – 75% of that occurring across

naked GFB regions.

of 38

17

Discussion

The traditional concept of glomerular ultrafiltration suggests that the individual layers of

the GFB act in series, and thus each layer makes an important contribution to the

composition of the final ultrafiltrate (5). The SPS lies downstream of the traditional

components of the filtration barrier (endothelial glycocalyx, endothelial fenestrae, GBM

and podocyte filtration slits), and this additional barrier is therefore likely to make an

important contribution to filtration by the glomerular filtration barrier. The high proportion

of the GFB that is covered by the SPS (20) ensures that the significant retention of

10kDa-weight molecules by the SPS, and the restriction to fluid flux imposed by the

SPS, are likely to be biologically significant effects. A second consequence of barriers

acting in series is that a change in the sieving properties of any one of the layers in the

barrier will have an impact on the overall sieving properties of the entire barrier (5).

Podocytes possess an actin-rich cytoskeleton (27), have stretch-activated ion channels

in their plasma membrane (19), exhibit changes in intracellular calcium in response to a

stimulus for tubuloglomerular feedback (23) and have a contractile phenotype in

response to physiological relevant stimuli (e.g. mechanical stress) in vitro (8, 10). If

these observations in vitro reflect an ability of the podocyte to alter its architecture in

vivo, then a change in the morphology of the SPS may be one mechanism by which a

change in podocyte cell shape influences the permeability characteristics of the

glomerular filtration barrier.

It has not been possible to distinguish the exact mechanism by which the SPS restricts

the movement of water and intermediate molecular weight solute. The dominant

of 38

18

resistance to molecular flux across the SPS is unlikely to reside at the entrance to the

space, since there is a large filtration surface area into each individual SPS (20), and

there do not appear to be any differences in the morphology of the underlying filtration

barrier in SPS-covered, as compared with naked, regions. Mathematical modelling

indicates that the sub-podocyte exit pores, whose infrequency determines that the area

that is available for efflux from the SPS is 25-fold lower than that available for influx into

each space (20), provide an important contribution to the hydraulic resistance of the

SPS (21). The importance of these exit pores on solute ultrafiltration remains to be

determined experimentally, but mathematical modelling (21) indicates that the SPS exit

pore may be the primary site of restriction to fluid flow. An alternative explanation to

restriction imposed by SPS exit pores is that the flux of molecules across the SPS cavity

may be the dominant form of restriction. The presence of glycocalyx, which appears to

line the interior of the SPS (11), may restrict solute and/or water movement (1, 33).

Finally, solute molecules may be selectively taken up by an active process within the

SPS, such as macropinocytosis or endocytosis (29, 35), although such phenomena

have not been demonstrated to occur over a period of seconds, as observed in our

studies.

The interpretations of the experimental results outlined above rely on a number of

assumptions. Firstly, tracer If was examined in regions of the glomerular capillary wall

that lay immediately beneath abluminal cell bodies or processes on the outer aspect of

peripheral capillary loops, which we have assumed to represent SPS. It was not

possible to identify SPS regions in living tissue with as ultra high resolution as is

of 38

19

possible with electron microscopy. However, we did combine a high temporal resolution

for physiological studies with simultaneous high spatial resolution images with the

multiphoton technique to give as much confidence as possible that we were measuring

flux in an SPS (Figs. 1-2, 7). Secondly, we have assumed that the rise in dextran If that

is observed in SPS regions during glomerular perfusion is not due to the plane of

imaging gradually shifting from a cell-rich wall region into a solute-rich capillary lumen.

Two major arguments can be raised against this possibility: dye If remained

approximately constant during the perfusion period, indicating that even if the plane of

imaging did move, then it did so into an equivalently cell-rich region; and the absence of

a similar rise in tracer If in naked capillary wall regions adjacent to the SPS in which

dextran If rose, or a similar effect for LY If in SPS regions.

Finally, the calculations of relative solvent flux through SPS-covered and naked GFB

regions require three assumptions. A. The permeability coefficients of the GFB with and

without SPS coverage are the same. Although there is no a priori reason to think that

this is not the case, the permeability values have not been determined. At the current

time there has been no systematic investigation of the ultrastructure of the GBM,

podocyte slit diaphragms or endothelial fenestrations underneath the SPS compared

with naked regions, but this needs to be carried out. B. The solute and solvent fluxes

were in steady state. The finding that LY If was stable shortly after initiation of perfusion

indicates a steady state condition. C. The height of the SPS that is used in the

calculation of Jv across SPS is representative. Conservative estimates of SPS height

were employed in the calculations of Jv across SPS (0.7 µm and 0.35 µm), as compared

of 38

20

with SPS height measured from the multiphoton images (mean 0.8µm), thus ensuring

that estimates of Jv across SPS are likely to be conservative in magnitude.

Anatomically distinct non-fluorescent regions bounded by the GFB and abluminal cell

bodies/processes, taken to represent SPS were observed less frequently with

multiphoton microscopy, as compared with electron microscopy (20). In addition, whilst

the mean SPS height observed with confocal microscopy is comparable with mean SPS

height under non-physiological perfusion conditions (20), multiphoton-derived heights

were significantly greater than those obtained with electron microscopy studies using

physiologically relevant perfusion conditions (21). It was not possible to identify most of

the narrow but extensive portions of the SPS that are so readily evident using electron

microscopy (see figure 1), and this lack of spatial resolution would account for the

apparent reduction in SPS size and frequency. Alternatively, SPS dimensions may vary

with the magnitude of Starling’s forces (20).

Our observations indicate two parallel pathways for solute and water movement across

the GFB; a higher resistance pathway through SPS-covered regions, and a relatively

low resistance pathway through areas devoid of SPS. The concept of heterogeneous

molecular flux across the GFB is consistent with those of Drumond & Deen (6). These

authors used mathematical modelling to demonstrate low rates of flux across portions of

the GBM immediately adjacent to endothelial cell or podocyte cytoplasm, and significant

hydrostatic pressure gradients acting within the GBM. Whilst the parallel pathway

implication of our results act on a different scale, they are consistent with the concept of

heterogeneous molecular movement across the GFB as proposed by Drumond & Deen

(6). Nevertheless, the observations reported in this and the companion paper need to

of 38

21

be reconciled with the plethora of physiological studies reporting GFB hydraulic

conductivity – area products (LpA) (3) and macromolecular sieving coefficients (θ) (4). In

terms of water flux, our observations indicate that the hydraulic resistance of the GFB

(glycocalyx-endowed fenestrated endothelial cells, GBM and podocyte filtration slits) is

lower than previously estimated, and that LpA values obtained from studies involving

micropuncture of glomerular capillaries and proximal tubules that were assumed to

represent GFB LpA are in fact based on the composite of fluid flux across a lower

resistance naked GFB pathway summated with the proportion of fluid flux that traverses

SPS. Significant retention of 10kDa dextran by SPS also indicates that the degree of

sieving that is imposed by the GFB (as defined above) is lower than previously

estimated, and that SPS make an important contribution to the composition of primary

ultrafiltrate. Whether SPS make a similar contribution to the sieving of macromolecules

of other sizes (e.g. albumin) is of interest.

It is not yet known whether the SPS is altered in disease states, or whether the

SPS can contribute to glomerular dysfunction. A number of conditions are associated

with proteinuria, and yet have minor, or minimal changes in the glomerular architecture,

on light microscopy and modest to moderate abnormalities on EM. These include

conditions such as hypertension, minimal change syndrome, and the early glomerular

lesions of diabetes (31). Closer examination of the SPS structure may reveal

abnormalities previously undescribed. Foot process effacement and extensive podocyte

architecture disruption may lead to increased macromolecular flux through the SPS,

leading to proteinuria. Alternatively an increase in resistance through the SPS may

of 38

22

result in hyperfiltration through uncovered regions. A dysfunction of this fourth layer of

the glomerular filtration barrier may therefore contribute to proteinuria.

In summary, we have shown, using an isolated perfused glomerulus and in vivo

multi-photon imaging technique, that the movement of the intermediate weight molecule

10kDa-dextran rhodamine was retarded in regions of the glomerular capillary wall that

are endowed with SPS. In contrast, the movement of the low molecular weight

substance lucifer yellow (≈ 450 Da) or 3kDa dextran was not retarded across SPS-

covered GFB. These observations indicate a size-selective barrier downstream of

podocyte foot processes. There was no retardation in the movement of either solute

across regions of the capillary wall that were devoid of SPS (“naked regions”). In

addition, substantially less solvent traversed SPS-covered GFB regions as compared

with naked GFB regions, and conservative estimates indicate that this reduction in Jv is

in the order of 66 – 75%. These observations indicate that the SPS does indeed impart

additional resistance to water movement across the GFB, as postulated by Neal et al

(20).

of 38

23

Acknowledgements

The authors would like to acknowledge support from the British Heart Foundation

(BB2000003), the Wellcome Trust (69134), the British Microcirculation Society, and the

National Institutes of Health (DK64324). I. Toma was a National Kidney Foundation

Postdoctoral Research Fellow, A. Sipos was an American Heart Association

Postdoctoral Research Fellow.

of 38

24

Figure 1. Imaging of SPS-covered and naked glomerular filtration barrier regions and

solute flux across them. A, 3D reconstruction from electron micrographs of a segment of

glomerular filtration barrier (GFB) (orange: endothelial cells; green: glomerular basement

membrane; blue: podocyte foot processes; yellow: sub-podocyte space; grey: podocyte cell

body). i, the GFB is not covered by SPS (“naked”) ii, An SPS lies between the traditional tri-

layered GFB and the podocyte cell body (“SPS-covered”);. Pink arrows represent fluid and

solute movement, which is directly from capillary lumen to peripheral urinary space in the naked

region (i), but must occur via a highly restricted and tortuous route in the SPS-covered region

(ii). B, representative multi-photon image of an isolated perfused glomerulus, loaded with 25 µM

quinacrine (labeling cell cytoplasm). Two rectangular areas are magnified on the right as

indicated. White arrows indicate regions of interest (i: naked region; ii: SPS-covered region).

The (ii) white arrows identify non-fluorescent regions under the podocyte cell body bounded by

plate-like extension from the abluminal cell and the glomerular capillary wall, representing a

sub-podocyte space. AA: afferent arteriole; BC: Bowman’s capsule; PUS: peripheral urinary

space. CB: cell body. CL: capillary lumen. MD: macula densa; JG: renin producing

juxtaglomerular cells; PB: perfusion bath; Bar is 15 µm.

Figure 2. Ex-vivo imaging of solute flux across SPS-covered and naked regions of the

glomerular filtration barrier. Representative images of an isolated and microperfused

glomerulus labeled with quinacrine (green) taken at the indicated time points after the start of

microperfusion with rhodamine B-conjugated 10kDa dextran (red). The solutes were perfused

for 1 min and washed out afterwards (60-64 s). Quinacrine intensely labeled cell bodies and

nuclei of podocytes, endothelial cells and mesangial cells. Unstained dark regions between the

filtration barrier and the podocyte body (SPS) are clearly visible (ii arrow). Appearance of

dextran (red) in glomerular capillaries and filtration into naked (I arrow) and SPS-covered

of 38

25

regions of the filtration barrier (ii arrow) was clearly visible. BC: Bowman’s capsule; PUS:

peripheral urinary space. CB: podocyte cell body. CL: capillary lumen. Bar is 20 µm.

Figure 3. 10kDa-dextran-rhodamine selectively accumulates in sub-podocyte spaces;

representative examples. A, naked GFB region. Fluorescence intensity per unit area (If) of

10kDa-dextran-rhodamine remains constant in the capillary lumen (�) during the period of

stable perfusion of isolated glomeruli (10-35 s), revealed by a near-flat regression line (broken

line) applied to these data points. dIf/dt (solid line) in naked GFB regions (without SPS cover) is

also approximately zero, indicating a lack of restriction of 10kDa-dextran after flux across naked

portions of the GFB. B, sub-podocyte space region. In contrast, If increases during the same

period of perfusion in the covered portion of GFB (SPS region; �, solid line) whilst remaining

constant in the capillary lumen (�, broken line), indicating accumulation of this intermediate-

weight molecule in the SPS.

Figure 4. Lucifer yellow does not accumulate in sub-podocyte spaces; representative

examples. Lucifer yellow If remained approximately constant in the capillary lumen (�, broken

line) and capillary wall (�, solid line) regions, irrespective of whether the capillary wall region

was identified as naked GFB (A) or as SPS-covered (B). This indicates freedom of movement

(i.e. no accumulation) of lucifer yellow across the GFB, irrespective of coverage by SPS. Mean

lucifer yellow fluorescence intensity (LY If) in naked GFB regions (nakIf) was 65% of that in

adjacent capillary lumen regions (lumIf). mean If in covered GFB regions was 46% of that in

adjacent capillary lumen regions. C. Naked GFB regions comprise only GFB and urinary space

(where LY If = lumIf), hence the percentage reduction in If indicates the percentage of GFB

volume from which LY is excluded (y: 1.27µm; yGFB: 1.02µm; percentage exclusion: 44%). D.

Covered GFB regions (y: 1.55µm) comprise GFB (44% of volume excludes LY: yGFB = 1.02µm),

of 38

26

SPS, and podocyte cell cytoplasm (100% of volume excludes LY). Assuming ySPS = 0.7µm (20),

the reduction in covIf/lumIf (as compared with nakIf/lumIf) cannot be explained by the anatomical

difference between these two regions (i.e. the presence of podocyte cell bodies). There must be

fewer LY molecules per unit volume of SPS as compared with the adjacent capillary lumen (and

hence with the urinary space in naked GFB regions), indicating reduced fluid flux across the

SPS (Jv is 66% of that occurring across naked urinary space regions). Assuming ySPS = 0.35µm

(21), the presence of podocyte cell bodies again cannot account for the reduction in covIf/lumIf as

compared with nakIf/lumIf,, and Jv is calculated to be 75% of that occurring across naked urinary

space regions.

Figure 5. Selective accumulation of 10kDa-dextran-rhodamine in SPS. The rate of change

of fluorescence intensity attributable to 10kDa-dextran in covered regions ([d(dextran If)/dt]SPS)

was significantly greater than that observed in adjacent capillary lumen regions ([d(dextran

If)/dt]lumen), resulting in consistently positive [d(dextran If)/dt]wall - [d(dextran If)/dt]lumen values (�,

p<0.0001, one-sample t-test), indicating accumulation of 10kDa-dextran-rhodamine in SPS.

This accumulation was not observed for 10kDa-dextran in naked GFB regions, nor for lucifer

yellow in either covered or naked GFB regions, (all other [d(tracer If)/dt]wall - [d(tracer If)/dt]lumen

values not significantly different from zero: one-sample t-test; and [d(tracer If)/dt]wall - [d(tracer

If)/dt]lumen significantly different from 10kDa-dextran-rhodamine in SPS: ∆, p<0.001, one-way

ANOVA with Bonferroni correction). N=6.

Figure 6. Ratiometric imaging of simultaneously perfused 10kDa and 3kDa dextrans. A,

naked GFB region. Fluorescence intensities per unit area (If) of both the 10kDa and 3kDa

dextrans remain constant in the Bowman’s space close to the GFB during the period of stable

perfusion of isolated glomeruli (60-90 s), revealed by a near-flat regression line applied to these

of 38

27

data points. This indicates a lack of restriction of both dextrans after flux across naked portions

of the GFB. B, sub-podocyte space region. In contrast, If of the 10kDa dextran increases during

the same period of perfusion in the covered portion of GFB whilst If of the 3kDa dextran remains

constant in the SPS. This indicates the accumulation of the 10kDa dextran, but not the 3kDa

dextran in the SPS. C, ratio of 10kDa and 3kDa dextrans. The ratio of 10kDa/3kDa dextrans

continuously rose above 1 in the SPS-covered region (solid line), but stayed constant around 1

in the naked regions (dotted line). Again, this indicates the relative accumulation of the 10kDa

compared to the 3kDa dextran in the SPS, but not in the naked regions of GFB.

Figure 7. In vivo imaging of solute flux across SPS-covered and naked regions of the

glomerular filtration barrier. Representative images of a glomerulus in the intact rat kidney

labeled with Hoechst 33342 (nuclear stain) and 500kDa dextran-fluorescein (plasma labeling)

(green). Unstained dark regions between the filtration barrier and the podocyte body (SPS) are

visible (ii arrows). Appearance of 10kDa dextran injected acutely in iv bolus (red) in glomerular

capillaries (at 2s) and its temporary accumulation in SPS-covered regions of the filtration barrier

was visible at 4s (inset, ii arrow), which appeared greater than in adjacent naked regions (i

arrow). BC: Bowman’s capsule; PUS: peripheral urinary space. CB: podocyte cell body. CL:

capillary lumen. Bar is 20 µm.

of 38

28

Table 1

[d(tracer If)/dt]wall - [d(tracer If)/dt]lumen n p-value (ANOVA)

10kDa-dextran-rhodamine,SPS 0.74 ± 0.12 units.s-1 13 <0.001

10kDa-dextran-rhodamine,naked -0.15 ± 0.12 units.s-1 6 n/s

Lucifer yellow, SPS -0.04 ± 0.13 units.s-1 12 n/s

Lucifer yellow, naked -0.07 ± 0.11 units.s-1 7 n/s

Table 1. 10kDa-dextran-rhodamine selectively accumulates in SPS-covered regions of the

glomerular filtration barrier.

of 38

29

References

1. Adamson RH. Permeability of frog mesenteric capillaries after partial pronase digestion of the endothelial glycocalyx. J Physiol 428: 1-13, 1990. 2. Benzing T. Signaling at the slit diaphragm. J Am Soc Nephrol 15: 1382-1391, 2004. 3. Brenner BM, Troy JL, and Daugharty TM. The dynamics of glomerular ultrafiltration in the rat. J Clin Invest 50: 1776-1780, 1971. 4. Chang RL, Ueki IF, Troy JL, Deen WM, Robertson CR, and Brenner BM. Permselectivity of the glomerular capillary wall to macromolecules. II. Experimental studies in rats using neutral dextran. Biophys J 15: 887-906, 1975. 5. Deen WM, Lazzara MJ, and Myers BD. Structural determinants of glomerular permeability. Am J Physiol Renal Physiol 281: F579-596, 2001. 6. Drumond MC and Deen WM. Structural determinants of glomerular hydraulic permeability. Am J Physiol 266: F1-12, 1994. 7. Ellis EN, Steffes MW, Chavers B, and Mauer SM. Observations of glomerular epithelial cell structure in patients with type I diabetes mellitus. Kidney Int 32: 736-741, 1987. 8. Endlich N, Kress KR, Reiser J, Uttenweiler D, Kriz W, Mundel P, and Endlich K. Podocytes respond to mechanical stress in vitro. J Am Soc Nephrol 12: 413-422, 2001. 9. Eremina V, Sood M, Haigh J, Nagy A, Lajoie G, Ferrara N, Gerber HP, Kikkawa Y, Miner JH, and Quaggin SE. Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases. J Clin Invest 111: 707-716, 2003. 10. Friedrich C, Endlich N, Kriz W, and Endlich K. Podocytes are sensitive to fluid shear stress in vitro. Am J Physiol Renal Physiol 291: F856-865, 2006. 11. Gattone VH, Pennington J, Miller C, Molitoris B, and Bacallao R. Use of High Pressure Freezing and Freeze Substitution Processing of Tissues. Experimental Biology 2006, Moscone Convention Center, San Francisco, CA, 2006, p. LB5. 12. Hall B. Studies of normal glomerular structure by electron microscopy. 5th American Conference on Nephrotic Syndrome, New York, 1954, p. 1-38. 13. Hjalmarsson C, Johansson BR, and Haraldsson B. Electron microscopic evaluation of the endothelial surface layer of glomerular capillaries. Microvasc Res 67: 9-17, 2004. 14. Jeansson M and Haraldsson B. Morphological and functional evidence for an important role of the endothelial cell glycocalyx in the glomerular barrier. Am J Physiol Renal Physiol 290: F111-116, 2006. 15. Kang JJ, Toma I, Sipos A, McCulloch F, and Peti-Peterdi J. Quantitative imaging of basic functions in renal (patho)physiology. Am J Physiol Renal Physiol 291: F495-502, 2006. 16. 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, and Tryggvason K. Positionally cloned gene for a novel glomerular

of 38

30

protein--nephrin--is mutated in congenital nephrotic syndrome. Mol Cell 1: 575-582, 1998. 17. Kriz W and Bankir L. A standard nomenclature for structures of the kidney. The Renal Commission of the International Union of Physiological Sciences (IUPS). Kidney Int 33: 1-7, 1988. 18. Levick JR and Smaje LH. An analysis of the permeability of a fenestra. Microvasc Res 33: 233-256, 1987. 19. Morton MJ, Hutchinson K, Mathieson PW, Witherden IR, Saleem MA, and Hunter M. Human podocytes possess a stretch-sensitive, Ca2+-activated K+ channel: potential implications for the control of glomerular filtration. J Am Soc Nephrol 15: 2981-2987, 2004. 20. Neal CR, Crook H, Bell E, Harper SJ, and Bates DO. Three-dimensional reconstruction of glomeruli by electron microscopy reveals a distinct restrictive urinary subpodocyte space. J Am Soc Nephrol 16: 1223-1235, 2005. 21. Neal CR, Muston PR, Njegovan D, R V, Harper SJ, Deen WM, and Bates DO. Glomerular filtration into the sub-podocyte space is highly restricted under physiological perfusion conditions. Companion paper, 2007. 22. Noakes PG, Miner JH, Gautam M, Cunningham JM, Sanes JR, and Merlie JP. The renal glomerulus of mice lacking s-laminin/laminin beta 2: nephrosis despite molecular compensation by laminin beta 1. Nat Genet 10: 400-406, 1995. 23. Peti-Peterdi J. Calcium wave of tubuloglomerular feedback. Am J Physiol Renal Physiol 291: F473-480, 2006. 24. Peti-Peterdi J. Multiphoton imaging of renal tissues in vitro. Am J Physiol Renal Physiol 288: F1079-1083, 2005. 25. Rodewald R and Karnovsky MJ. Porous substructure of the glomerular slit diaphragm in the rat and mouse. J Cell Biol 60: 423-433, 1974. 26. Rostgaard J and Qvortrup K. Electron microscopic demonstrations of filamentous molecular sieve plugs in capillary fenestrae. Microvasc Res 53: 1-13, 1997. 27. Saleem MA, Ni L, Witherden I, Tryggvason K, Ruotsalainen V, Mundel P, and Mathieson PW. Co-localization of nephrin, podocin, and the actin cytoskeleton: evidence for a role in podocyte foot process formation. Am J Pathol 161: 1459-1466, 2002. 28. Salmon AH, Neal CR, Bates DO, and Harper SJ. Vascular endothelial growth factor increases the ultrafiltration coefficient in isolated intact Wistar rat glomeruli. JPhysiol 570: 141-156, 2006. 29. Shaw AS. Handling of Protein by the Podocyte. American Society of Nephrology Renal Week 2006, San Diego, CA, USA, 2006. 30. Thorner PS. Alport syndrome and thin basement membrane nephropathy. Nephron Clin Pract 106: c82-88, 2007. 31. Tomson CRV. Cardiovascular disease in chronic renal failure. In: Comprehensive Clinical Nephrology, edited by Johnson RH and Feehally J: Mosby, 2000, p. 70.71-70.14. 32. Venturoli D and Rippe B. Ficoll and dextran vs. globular proteins as probes for testing glomerular permselectivity: effects of molecular size, shape, charge, and deformability. Am J Physiol Renal Physiol 288: F605-613, 2005.

of 38

31

33. Vink H and Duling BR. Capillary endothelial surface layer selectively reduces plasma solute distribution volume. Am J Physiol Heart Circ Physiol 278: H285-289, 2000. 34. Wartiovaara J, Ofverstedt LG, Khoshnoodi J, Zhang J, Makela E, Sandin S, Ruotsalainen V, Cheng RH, Jalanko H, Skoglund U, and Tryggvason K. Nephrin strands contribute to a porous slit diaphragm scaffold as revealed by electron tomography. J Clin Invest 114: 1475-1483, 2004. 35. Welsch T, Endlich N, Gokce G, Doroshenko E, Simpson JC, Kriz W, Shaw AS, and Endlich K. Association of CD2AP with dynamic actin on vesicles in podocytes. Am J Physiol Renal Physiol 289: F1134-1143, 2005. 36. Yamada E. The fine structure of the renal glomerulus of the mouse. J Biophys Biochem Cytol 1: 551-566, 1955.

of 38

32

Ai ii

B

iiAA

BC

CL

CB

PB

PUSMD

i

ii

CL

PUS

CB

CB

CL

Figure 1

JG

Ai ii

B

iiAA

BC

CL

CB

PB

PUSMD

i

ii

CL

PUS

CB

CB

CL

Figure 1

JG

of 38

33

BC

CL CB

PUSCB

CL

0s 2s 4s

60s 62s 64s

Figure 2

i

ii

BC

CL CB

PUSCB

CL

0s 2s 4s

60s 62s 64s

Figure 2

i

ii

of 38

34

0102030405060708090

100

time (s)

I f(a

rbitr

ary

units

)

0 10 20 30 40 505 15 25 35 45

A 10kDa-dextran, naked capillary wall

capillary lumennaked capillary wall

0

10

20

30

40

50

60

70

80

0 5 10 15 20 25 30 35 40 45 50time (s)

I f(a

rbitr

ary

units

)

B10kDa-dextran, sub-podocyte space

capillary lumensub-podocyte space

Figure 3

0102030405060708090

100

time (s)

I f(a

rbitr

ary

units

)

0 10 20 30 40 505 15 25 35 45


capillary lumennaked capillary wall

0102030405060708090

100

time (s)

I f(a

rbitr

ary

units

)

0 10 20 30 40 505 15 25 35 450

102030405060708090

100

0102030405060708090

100

time (s)

I f(a

rbitr

ary

units

)

0 10 20 30 40 505 15 25 35 450 10 20 30 40 500 10 20 30 40 505 15 25 35 45


capillary lumennaked capillary wallcapillary lumennaked capillary wall

0

10

20

30

40

50

60

70

80

0 5 10 15 20 25 30 35 40 45 50time (s)

I f(a

rbitr

ary

units

)

B10kDa-dextran, sub-podocyte space

capillary lumensub-podocyte spacecapillary lumensub-podocyte space

Figure 3

of 38

35

0

20

40

60

80

100

120

0 10 20 30 40 50time (s)

B lucifer yellow, sub-podocyte space capillary lumensub-podocyte space

I f(a

rbitr

ary

units

)

0

20

40

60

80

100

120

0 10 20 30 40 50time (s)

A lucifer yellow, naked capillary wallcapillary lumennaked capillary wall

I f(a

rbitr

ary

units

)

Figure 4

lumIf

covIf

mean nakIf / lumIf = 0.65±0.07 mean covIf / lumIf = 0.46±0.06

C D

zx

yGFB

yUS

y

zx

yGFB

ySPSyCB

y

sub-podocyte spacenaked capillary wall

0

20

40

60

80

100

120

0 10 20 30 40 50time (s)

B lucifer yellow, sub-podocyte space capillary lumensub-podocyte spacecapillary lumensub-podocyte space

I f(a

rbitr

ary

units

)

0

20

40

60

80

100

120

0 10 20 30 40 50time (s)

A lucifer yellow, naked capillary wallcapillary lumennaked capillary wallcapillary lumennaked capillary wall

I f(a

rbitr

ary

units

)

Figure 4

lumIf

covIf

lumIf

covIf

mean nakIf / lumIf = 0.65±0.07 mean covIf / lumIf = 0.46±0.06mean nakIf / lumIf = 0.65±0.07 mean covIf / lumIf = 0.46±0.06

C D

zx

yGFB

yUS

y

zx

yGFB

yUS

y

zx

yGFB

ySPSyCB

y

zx

yGFB

ySPSyCB

y

sub-podocyte spacenaked capillary wall

of 38

36

SPS naked SPS naked

10kDa-dextran-rhodamine lucifer yellow

[d(tr

acer

I f)/d

t]wal

l -[d

(trac

erI f)

/dt]lu

men

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0 •∆

Figure 5

SPS naked SPS naked

10kDa-dextran-rhodamine lucifer yellow

[d(tr

acer

I f)/d

t]wal

l -[d

(trac

erI f)

/dt]lu

men

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0 •∆

SPS nakedSPSSPS nakednaked SPS nakedSPSSPS nakednaked

10kDa-dextran-rhodamine10kDa-dextran-rhodamine lucifer yellowlucifer yellow

[d(tr

acer

I f)/d

t]wal

l -[d

(trac

erI f)

/dt]lu

men

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0 •∆

Figure 5

of 38

37

Naked region

0

100

200

300

400

500

600

700

0 20 40 60 80 100 120 140 160

3kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If)

SPS covered region

0

200

400

600

800

1000

1200

1400

0 20 40 60 80 100 120 140 160

3kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If)

00.20.40.60.8

11.21.41.61.8

2

0 20 40 60 80 100 120 140 160

NakedCovered

Rat

io10

kDa:

3kD

adex

tran

A

B

C

Figure 6

Time (sec)

Time (sec)

Time (sec)

Naked region

0

100

200

300

400

500

600

700

0 20 40 60 80 100 120 140 160

3kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If) Naked region

0

100

200

300

400

500

600

700

0

100

200

300

400

500

600

700

0 20 40 60 80 100 120 140 160

3kDa10kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If)

SPS covered region

0

200

400

600

800

1000

1200

1400

0 20 40 60 80 100 120 140 160

3kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If)

SPS covered region

0

200

400

600

800

1000

1200

1400

0 20 40 60 80 100 120 140 160

3kDa10kDa10kDa

Fluo

resc

ence

inte

nsity

abov

eba

selin

e(∆

If)

00.20.40.60.8

11.21.41.61.8

2

0 20 40 60 80 100 120 140 160

NakedCovered

Rat

io10

kDa:

3kD

adex

tran

A

B

C

Figure 6

Time (sec)

Time (sec)

Time (sec)

of 38

38

BCPUS

CB

CL

0s 2s 4s

Figure 7

iii

ii

ii

BCPUS

CB

CL

0s 2s 4s

Figure 7

iii

ii

ii

of 38

Evidence for restriction of fluid and solute movement across the glomerular capillary wall by the...

Documents

Transcript of Evidence for restriction of fluid and solute movement across the glomerular capillary wall by the...