Fluid flow in the juxtaglomerular interstitium visualized in vivo

【摘要】  Earlier electron microscopy studies demonstrated morphological signs of fluid flow in the juxtaglomerular apparatus (JGA), including fenestrations of the afferent arteriole (AA) endothelium facing renin granular cells. We aimed to directly visualize fluid flow in the JGA, the putative function of the fenestrated endothelium, using intravital multiphoton microscopy of Munich-Wistar rats and C57BL6 mice. Renin content of the AA correlated strongly with the length of the fenestrated, filtering AA segment. Fluorescence of the extracellular fluid marker lucifer yellow (LY) injected into the cannulated femoral vein in bolus was followed in the renal cortex by real-time imaging. LY was detected in the interstitium around the JG AA before the plasma LY filtered into Bowman's capsule and early proximal tubule. The fluorescence intensity of LY in the JGA interstitium was 17.9 ± 3.5% of that in the AA plasma ( n = 6). The JGA fluid flow was oscillatory, consisting of two components: a fast (one every 5-10 s) and a slow (one every 45-50 s) oscillation, most likely due to the rapid transmission of both the myogenic and tubuloglomerular feedback (TGF)-mediated hemodynamic changes. LY was also detected in the distal tubular lumen about 2-5 s later than in the AA, indicating the flow of JGA interstitial fluid through the macula densa. In the isolated microperfused JGA, blocking the early proximal tubule with a micropipette caused significant increases in MD cell volume by 62 ± 4% ( n = 4) and induced dilation of the intercellular lateral spaces. In summary, significant and dynamic fluid flow exists in the JGA which may help filter the released renin into the renal interstitium (endocrine function). It may also modulate TGF and renin signals in the JGA (hemodynamic function).

【关键词】  intravital microscopy fenestration shortloop feedback quinacrine lucifer yellow

THE JUXTAGLOMERULAR APPARATUS (JGA) is a major structural component of the renin-angiotensin system and is an important regulatory site of renal salt and water conservation and systemic blood pressure homeostasis. Major functions of the JGA include the feedback regulation of glomerular filtration rate (tubuloglomerular feedback, TGF) and renin release from JG granular cells. These mechanisms are based on the salt content of the distal tubule, as well as the anatomical and functional connection between macula densa (MD) cells in the distal tubule and the glomerular vascular pole of the same nephron ( 34 ).

One of the unique morphological characteristics of the JGA is the absence of capillaries and lymph vessels in the extraglomerular mesangium ( 2 ), the area between the afferent (AA) and efferent arterioles (EA), the glomerulus, and the MD. This avascular area consists of densely packed mesangial cells amidst an interstitium filled with basal laminar material ( 2, 15 ). Based on these observations, it was presumed that the JGA interstitial fluid is sequestered from the rest of the cortical interstitium, permitting the localized accumulation of electrolytes ( 18 ) as well as humoral mediators (such as ATP, adenosine, PGE 2, and NO) underneath the MD ( 4, 22, 33 ). All of these morphological features are thought to create an ideal milieu for humoral cell-to-cell communication, which is an integral part of TGF and renin release mechanisms ( 2, 15, 32 ).

In contrast, earlier electron microscopy studies demonstrated morphological prerequisites and signs of bulk fluid flow in the JGA. These include 1 ) endothelial fenestrations in the terminal, renin-positive AA segments (25-30), 2 ) the presence of Bowman's space and podocyte foot processes in the JGA facing the extraglomerular mesangial interstitium ( 25 ), and 3 ) the demonstration of high endothelial permeability in the JGA using ferritin particles ( 26, 27 ). These studies suggested that the fluid composition and balance in the JGA (and consequently the TGF and renin release mechanisms) do not depend solely on the reabsorptive transport of the MD, but also on the rate of bulk fluid flow directed to the JGA interstitium ( 25, 27, 30 ). The existence of significant fluid flow, however, has been difficult to study due to the small size and inaccessibility of the JGA.

Multiphoton excitation fluorescence microscopy offers a state-of-the-art imaging technique for the visualization of dynamic renal processes in real-time ( 7, 10, 14, 19 - 21 ). Noninvasive, quantitative imaging of basic functions in the intact kidney has been recently reported using this technology ( 10, 39 ). The processes of fluid filtration through the glomerular barrier, glomerular permeability, tubular fluid flow, and capillary blood flow have been visualized and quantified with high optical and temporal resolution ( 10, 14, 39 ).

The aim of the present study was to directly visualize the putative fluid flow in the JGA using in vivo multiphoton imaging. Both morphological and functional signs of a significant and dynamic interstitial fluid flow are presented here, challenging the existing paradigm of the stable and isolated JGA environment.

MATERIALS AND METHODS

Multiphoton excitation laser-scanning fluorescence microscopy. The multiphoton microscope used in these studies was a Leica TCS SP2 AOBS MP confocal microscope system (Leica Microsystems, Heidelberg, Germany) with the following components: a Leica DM IRE2 inverted microscope powered by a wideband, fully automated, infrared (710-920 nm) combined photo-diode pump laser and mode-locked titanium: sapphire laser (Mai-Tai, Spectra-Physics, Mountain View, CA). Images were collected in time (xyt) series (0.5-5 Hz) with the Leica Confocal Software (LCS 2.61.1537) and analyzed with the LCS 3D, Process, and Quantify packages. The principles and advantages of conventional confocal and two-photon microscopy and their application in imaging renal tissues have been recently reviewed ( 14, 21 ).

Nontoxic, water-soluble fluorophores were injected intravenously to label specific structures in the living kidney as described before ( 10 ). Briefly, a 70-kDa dextran-rhodamine B conjugate (Invitrogen) was used to label the circulating plasma or intravascular space. Tubular segments and more specifically the content of individual renin granules were visualized using quinacrine ( 1, 10, 20, 21 ). The extracellular fluid marker lucifer yellow (LY, Invitrogen) was used to visualize interstitial fluid movements ( 6 ), since LY is impermeable to cell membranes but freely filtered through basement membranes ( 10 ) due to its small size (0.45 kDa). All three of the fluorescent probes were excited using the same, single excitation wavelength of 860 nm (Mai-Tai, Spectra-Physics), and the emitted, nondescanned fluorescent light was detected by two external photomultipliers (green and red channels) with the help of a FITC/TRITC filter block (Leica).

Animals. All animal protocols have been approved by the Institutional Animal Care and Use Committee at the University of Southern California. Munich-Wistar male rats (200 g, Harlan, Madison, WI) were maintained on standard diet, while C57BL6 mice (20 g, inbred) were fed either a standard or a low-salt (TD 90228: 0.01% NaCl, Harlan) rodent diet for up to 1 wk. Animals were anesthetized with thiobutabarbital (Inactin, 130 mg/kg body wt) alone (for rats) or in combination with 50 mg/kg body wt ketamine (for mice). 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-to 15-mm dorsal incision was made under sterile conditions and the kidney was exteriorized. To image the JGA in the mouse kidney, a less than 1-mm-thin cortical slice was removed (parenchymal window) to allow imaging of the most superficial AA and glomeruli. Bleeding was minimal and stopped spontaneously within two min. Imaging through the thin blood clot did not cause any apparent changes in image quality or in the characteristics of fluorescent probes. This procedure was not required for rats, since Munich-Wistar rats have superficial glomeruli. 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 as described by Dunn et al. ( 7 ) 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.

During in vivo imaging, systemic blood pressure of animals was monitored through a cannula inserted into the left femoral artery and using an analog single-channel transducer signal conditioner model BP-1 (World Precision Instruments, Sarasota, FL), transducer model BLPR. Calibration was performed using a Pressure manometer model PM-015 and data were collected using data-acquisition system QUAD-161.

In vitro isolated and microperfused AA-JGA-glomerulus. A superficial AA with its glomerulus and attached distal tubule containing the MD was microdissected from kidneys of New Zealand white female rabbits (500 g, Irish Farms, Norco, CA) and microperfused as described before ( 21 ). Briefly, the dissection medium was prepared from DMEM (DME mixture F-12; Sigma) with the addition of 1.2 g/l NaHCO 3, and 3% fetal bovine serum (Hyclone). Before use, this solution was aerated with 95% O 2 -5% CO 2 for 45 min, and the pH was adjusted to 7.4. The arteriole was perfused with a modified Krebs-Ringer-HCO 3 buffer containing (in mM) 115 NaCl, 5 KCl, 25 NaHCO 3, 0.96 NaH 2 PO 4, 0.24 Na 2 HPO 4, 1.2 MgSO 4, 2 CaCl 2, 5.5 D -glucose, 100 µM L -arginine, and perfusion pressure was maintained at around 50 mmHg (1 psi) throughout the experiment. The tubular perfusate was an isosmotic, low NaCl containing Ringer solution consisting of (in mM): 10 NaCl, 135 N -methyl- D -glucamine (NMDG) cyclamate, 5 KCl, 1 MgSO 4, 1.6 Na 2 HPO 4, 0.4 NaH 2 PO 4, 1.5 CaCl 2, 5 D -glucose, and 10 HEPES. The tubule segment was cannulated and perfused at a baseline rate of 2 nl/min. The bath was identical to the arteriolar perfusate and was continuously aerated with 95% O 2 -5% CO 2 and exchanged at a rate of 1 ml/min. The preparation was kept in the dissection solution and temperature was kept at 4°C until cannulation of the arteriole and tubule was completed then gradually raised to 37°C for the remainder of the experiment. Fluorescence imaging of MD cells was performed using Fluo-4 and Fura-Red, and a Leica TCS SP2 AOBS MP confocal microscope as described before ( 23 ). Chemicals, if not indicated, were purchased from Sigma (St. Louis, MO).

Data analysis. Data are expressed as means ± SE. Statistical significance was tested using paired t -test and linear regression. Significance was accepted at P < 0.05.

RESULTS

Endothelial fenestration. High-resolution images of the AA endothelium and JG renin cells were obtained from the rat and mouse kidney in vivo. The content of renin granules in JG cells was labeled using quinacrine, a selective marker of acidic and secretory granules, and the endothelium was stained using LY endocytosed from the plasma ( Fig. 1 ). Quinacrine staining of individual renin granules was highly specific and localized to the JG segment of the AA. As expected, the length of the quinacrine (renin)-labeled AA segment increased with respect to the time on salt-deficient diet (from around 20 up to 90 µm, Fig. 2 x -axis). Interestingly, LY stained the proximal and JG AA endothelium segments differently. In the proximal portion, the endothelium appeared as a solid and intensely fluorescent line, but it was weakly fluorescent with a discontinuous line pattern in the JG segment ( Fig. 1 A ). High magnification revealed a porous, fenestrated structure of AA endothelium adjacent to renin granules, but a thick, and smooth endothelium in renin-negative segments ( Fig. 1 B ). In fact, there was a highly significant correlation ( r = 0.91, n = 15) between the length of quinacrine-labeled, renin-positive AA (stimulated by salt-deficient diet) and that of the weakly LY fluorescent, fenestrated endothelium ( Fig. 2 ).

Fig. 1. Low ( A )- and high ( B )-magnification multiphoton image of the proximal (P) and juxtaglomerular (JG) segments of the afferent arteriole (AA) in mouse. Content of individual renin granules is labeled by quinacrine (green). The vascular endothelium is labeled by the endocytosis of lucifer yellow (yellow). Note the intense and continuous yellow labeling in the proximal AA segment compared with the weakly fluorescent dash-dot pattern in the JG, renin-positive part. G, glomerulus. Bar = 10 µm.

Fig. 2. Correlation between the lengths of renin-positive and fenestrated endothelial segments of the mouse AA. To stimulate renin content, animals were fed a salt-deficient diet for various times, up to 1 wk.

JGA fluid flow in vivo. A single intravenous bolus of the fluid marker LY was injected into the femoral vein, which appeared in the AA and glomerulus within 5 s and then filtered into the Bowman's space and early proximal tubule. A supplementary video file showing the appearance of LY in the JGA area is available in the online version of this article. Using high temporal resolution, fluorescence intensity changes of LY were measured in different regions of interest in the JGA as shown in Fig. 3. Simultaneously with the appearance of LY in the lumen of the AA, fluorescence of LY was detected in the JGA interstitium as well, around the AA ( Fig. 3 A ). The increase in LY fluorescence in the JGA interstitium preceded the appearance of LY in the Bowman's capsule and early proximal tubule ( Fig. 3, A and B ). A wave of interstitial LY fluorescence was observed moving away from the JGA in all directions (see the representative video supplement) with a speed of 27.9 ± 7.2 µm/s ( n = 6). Parallel with its plasma and glomerular clearance, LY quickly washed out of the JGA interstium within 5 s ( Fig. 3, A-D ). The maximal fluorescence intensity of LY in the JGA interstitium was 17.9 ± 3.5% of that in the AA plasma ( n = 6; Fig. 3 D ).

Fig. 3. In vivo imaging of the JG AA in the intact rat kidney. Circulating plasma is labeled with rhodamine B-conjugated 70-kDa dextran (red), content of individual renin granules with quinacrine (green). Note the significant renin content around the JG portion of the afferent (AA), but not around the proximal AA segment (P) or the efferent arteriole (EA). G, glomerulus. A-C : fluorescence of iv injected LY (yellow) was visualized in the AA and surrounding interstitium in real time. Panels show the time it first appeared in the AA and JGA ( A ), 2 s later ( B ) when it filtered into the early proximal tubule (PT), and another 4 s later ( C ) when it almost completely cleared from the JGA area. Scale is 20 µm. Note a typical thin-walled peritubular capillary in the top left corner surrounded by interstitial cells around the PT. For comparison, the AA and EA are at least twice the diameter and have thick vascular smooth muscle (or JG) cells in the vessel wall. The "dark structures" excluding LY are the many cell types in the particular optical section. Since the cell membranes are impermeable to LY, only the extracellular fluid is labeled, but cells are unstained. D : representative time-lapse changes in LY fluorescence in the AA lumen, JGA interstitium, and the PT. E : profile of LY fluorescence along a line segment in the JGA interstitium, parallel with the AA and upstream from JG toward P segments (as depicted in A ). The quinacrine-labeled JG AA portion is labeled as renin-segment. F : correlation between the length of quinacrine (renin)-labeled JG AA and that of the intensely LY-stained (LY-rich) segment of the JGA interstitium.

The distribution of LY fluorescence in the JGA interstitium was heterogeneous along the AA ( Fig. 3 A ). Only the JG portion closest to the glomerular vascular pole showed high LY intensity, while fluorescence around proximal AA segments was minimal ( Fig. 3, A and B ). The length of this intensely LY fluorescent segment of JGA interstitium showed significant correlation ( r = 0.85, n = 9) with the length of the quinacrine-labeled, renin-positive AA ( Fig. 3, E and F ).

In addition to LY, the presence of a high molecular weight dextran-rhodamine conjugate, originating from the circulating plasma, was also detected in the JGA interstitium ( Figs. 3 - 5 ). Interestingly, the fluorescence intensity of dextran-rhodamine in the JGA was oscillatory, similar to that in the Bowman's capsule ( Fig. 4 ). The oscillations showed the same phase and cycle time at these two locations, and had two components: a slow (one every 45-50 s) and a fast (one every 5-10 s) oscillation ( Fig. 4 ).

Fig. 5. In vivo imaging of the macula densa (MD)-containing JGA in the intact rat kidney after iv injection of LY ( A ). Circulating plasma is labeled with rhodamine B-conjugated 70-kDa dextran (red), content of individual renin granules with quinacrine (green). High magnification of afferent (AA) and efferent (EA) arterioles, G, and cortical thick ascending limb (cTAL) is shown. Scale is 20 µm. B : LY fluorescence in different regions of interest after iv injection. Magnitude of changes is not drawn to scale. Simultaneous measurements are shown (the vertical dotted line represents time 0 of LY injection), so the delay in the appearance of LY in the MD lumen is easy to read.

Fig. 4. Oscillations in the fluorescence intensity of a 70-kDa dextran-rhodamine B conjugate in the JGA interstitium and Bowman's capsule (BC). Magnitude of changes is not drawn to scale. Simultaneous measurements are shown (the vertical dotted line represents time 0 of imaging), so the phase and cycle time of oscillations are easy to read. Note that 2 types of oscillations are present: a slow component (one in every 45-50 s) and a fast component (one in every 5-10 s).

Following iv bolus injections, LY fluorescence was detected not only in the JGA interstitium, but in the tubular lumen of the MD as well ( Fig. 5 ). Luminal LY fluorescence appeared to be specific for the MD area and was not visible in slightly upstream segments of the thick ascending limb ( Fig. 5 A ). Maximal LY intensity in the MD lumen was observed about 2-5 s later than that in the AA and JGA ( Fig. 5 B ). The intratubular transit time of LY filtered in the glomerulus and transported to the MD area along the nephron was much longer, usually between 30-50 s (not shown). Systemic blood pressure was normal in all animals studied; the average mean arterial blood pressure was 98.6 ± 8.0 mmHg ( n = 6).

JGA fluid flow in vitro. Further experiments were performed to study fluid flow in the JGA, particularly in the MD area using the isolated, simultaneously microperfused AA-thick ascending limb with attached glomerulus preparation. The effect of increased Bowman's capsule hydrostatic pressure was visualized by blocking the outflow of glomerular filtrate with a holding pipette ( Fig. 6, and a supplementary video file is available at http://ajprenal.physiology.org/cgi/content/full/ ). In response to compressing the early proximal tubule with a micropipette (during constant AA perfusion with 50 mmHg), significant increases in the volumes of the Bowman's capsule, JGA interstitium, and the MD were observed ( Fig. 6, A and B ). MD cells increased their volume by 62 ± 4% ( n = 4), as measured by changes in intracellular fluorescence of fluo-4 ( Fig. 6 C ). Similar data were also obtained by using fura red (not shown). In addition to MD cell volume changes, the extraglomerular mesangial and MD intercellular lateral spaces also distended significantly, particularly in the center of the MD plaque ( Fig. 6 B ). All these responses were fully reversible upon removing the proximal tubule block ( Fig. 6 C, and video supplement).

Fig. 6. Differential interference contrast image of the in vitro microperfused, isolated JGA before ( A ) and after ( B ) the occlusion of the proximal tubule lumen. Note the significant increase in the BC and MD volume. Scale is 20 µm. C : representative recording of the changes in MD fluo-4 fluorescence (F 0 /F, normalized) during the occlusion.

DISCUSSION

In the present study, fluid flow in the JGA interstitium was directly visualized for the first time in the intact rat kidney using intravital multiphoton imaging. Both morphological prerequisites and functional signs of significant JGA interstitial fluid flow were observed. The length of the porous, fenestrated AA endothelium that is unique to JG segments correlated highly with the length of the quinacrine-labeled, renin-positive AA. Real-time imaging of the iv injected low molecular weight fluid marker LY revealed that a significant portion of plasma is filtered to the JGA interstitium through this renin-positive AA segment. Fluid flow into the JGA appeared to be very dynamic, changing rapidly in response to variations in JGA function and hemodynamics (TGF). Fluid flow was also observed in the MD region. Overall, these findings challenge the existing view of the JGA interstitium as a sequestered, stable environment.

The morphological features of the AA visualized in vivo in the present experiments are consistent with earlier works (25-30, 37) describing the heterogeneity of the AA using electron microscopy. In vivo labeling of the vascular endothelium by endocytosed LY ( Fig. 1 ) confirmed that the AA is not a vessel with uniform characteristics along its entire length. The two structurally and functionally different portions include a renin-negative proximal portion lined by continuous, nonpermeable endothelium and a renin-positive JG portion lined by fenestrated, highly permeable endothelium. It has been demonstrated that the lengths of these two segments are highly variable, tracking the actual status and activity of the renin-angiotensin system ( 27 ). Consistent with this, we found that salt-deficient diet increased the length of both the renin-containing and fenestrated endothelial AA segments ( Figs. 1 and 2 ). Also, there was a strong correlation between fenestration and LY permeability ( Fig. 3 F ).

Only a limited number of vascular beds feature fenestrated endothelium without a diaphragm, such as those in the glomerulus and endocrine tissues ( 8, 36 ). The presence of fenestrations in the JG portion of AA endothelium which was first described by Rosivall and Taugner ( 25 ) is very peculiar since intravascular pressure in this region is high (around 50-60 mmHg). This high pressure most likely favors significant fluid filtration into the interstitium. Fenestration of the JG AA was found in various species from amphibians to mammals (25-29), suggesting it has an important, general physiological function. It is very likely that fenestration is an integral part of the endocrine function of the JGA. It may help the diffusion of released renin directly into the circulating plasma through the endothelial fenestrae ( 12 ) and it could also help to filter renin into the cortical interstitium ( 20 ) along with the fluid it generated. Although it is a very active area of research, it is unclear at present how the renin content signals the formation of fenestrae, how these pores are formed, or how their size is controlled. Several studies suggest that vascular endothelial growth factor (VEGF), a major regulator of vascular permeability, is one of the key players in this phenomenon ( 8, 24, 31, 36 ).

It should be noted that the fenestrations appear to be larger in the intact living kidney when fluid actually moves through the pores, compared with that measured by TEM on fixed, dehydrated tissue sections. We believe that this is due to the different techniques used. On a similar note, the volume of the renal interstitium also seems to be larger in vivo. Histology textbooks describe the extraglomerular mesangium as "densely packed mesangial cells separated by long, narrow interstitial space filled with basal laminar material" ( 2, 15 ). This is not the case in the intact living kidney, where mesangial intercellular spaces are much wider ( Fig. 3 A ). These observations further support the existence of fluid filtration from the JG AA and the bulk fluid flow in the JGA.

As a noninvasive alternative to existing methods, multiphoton in vivo imaging can be used for the quantitative visualization of basic renal functions ( 10, 14, 39 ). In a recent innovative method ( 10 ) LY, a widely used extracellular fluid marker ( 6 ), was a superior tool to measure glomerular filtration and tubular flow rate due to its low molecular weight, excellent water solubility, and intense fluorescence. Compared with the demanding conventional micropuncture methods ( 38 ), this technique monitors the fluorescence of a single iv bolus of LY in real-time and noninvasively, so tubular flow, blood flow, and macula densa function are intact. In the present study, LY was useful to visualize fluid filtration from the AA lumen to the JGA interstitium with high temporal resolution ( Fig. 3, and video supplement). The LY iv bolus appeared in the lumen of AA within 5 s, and LY was detected around the AA and in the extraglomerular mesangium before the remainder of the circulating LY filtered into the Bowman's capsule, early proximal tubule, or appeared in peritubular capillaries ( Fig. 3, A-D ). LY fluorescence was detected in both the extraglomerular and intraglomerular mesangium ( Fig. 3 A ). This is consistent with the anatomical continuity between these two areas, and also with the fenestrated and filtering intraglomerular capillaries ( 25 ). The movement of LY in the renal interstitium was most likely due to both fluid filtration and diffusion. However, the fact that LY appeared in the JGA interstitium before it was filtered in the glomerulus is the unequivocal evidence of bulk fluid flow rather than the movement of LY in the interstitium. Based on LY fluorescence, fluid filtration from the AA was specific for the quinacrine-labeled, renin-positive, and fenestrated JG portion of AA ( Fig. 3, A-F ). In contrast, low LY intensity was detected in the interstitium around the proximal, upstream segment of AA ( Fig. 3, B and E ). Interestingly, the maximal LY fluorescence in the JGA interstitium was 15-20% of that in the AA plasma ( Fig. 3 D ). This is a very significant amount and supports the existence of an intense fluid filtration from the lumen of JG AA. It should be mentioned however, that the amount of interstitial LY probably has three sources: 1 ) filtration from the JG AA, 2 ) filtration from the Bowman's capsule which is bulging back into the extraglomerular mesangium ( 25, 29 ), and 3 ) from peritubular capillaries. However, based on the JGA intensity profile ( Fig. 3, A and E ) which shows the highest LY fluorescence adjacent to the JG AA and not at the glomerular entrance (G on Fig. 3 A ), the first of the three mechanisms, filtration from the JG AA, is certainly the most significant.

Filtration of plasma from the JG AA into the JGA interstitium and the fluid flow it generated responded rapidly to hemodynamic changes. Not only was the LY bolus cleared from the JGA interstitium within 4-5 s ( Fig. 3, C-D ), but fluorescence of a permeable plasma marker showed the same, phase-matched oscillations in the JGA-extraglomerular mesangium as those in the Bowman's space ( Fig. 4 ). As has been recently reported, the present experimental technique allows real-time measurements of both the myogenic and TGF mechanism-induced oscillations in glomerular and tubular functions ( 10 ). The myogenic mechanism produces fast (6-to 10-s periods), while the TGF produces slow (20-to 50-s periods) oscillations ( 9, 13 ). Figure 4 demonstrates that both of these mechanisms influence fluid filtration into the JGA interstitium and consequently the composition of JGA interstitial fluid. Even the fast myogenic oscillations were transmitted to the JGA fluid flow, as demonstrated by the oscillating fluorescence intensity of the marker in the JGA interstitium ( Fig. 4 ). The supplementary LY video shows only 10 s, during which time the JGA geometry does not change significantly. However, imaging for longer time periods (5 min, Fig. 4 ) reveals TGF-mediated oscillations in both fluorescence intensity and morphology of the extraglomerular mesangium. For further clarification, it is helpful to refer to another video of TGF oscillations in the intact kidney in Ref. 10. Both the volume and fluorescence intensity of the extraglomerular mesangium (red) clearly oscillates with TGF, as evidence of JGA fluid flow described in the present studies.

Interestingly, significant fluid flow was also directed to the MD area. Only a few seconds after the appearance of LY in the JG AA lumen, LY fluorescence was observed over the cross section of the MD plaque and in the tubular lumen ( Fig. 5, A and B ). The luminal LY signal was specific for the MD area, since upstream segments of the thick ascending limb showed only minimal fluorescence ( Fig. 5 A ). LY most likely diffused or filtered through the MD epithelium into the tubular lumen via the intercellular lateral spaces, because the cell membrane is impermeable to LY (Molecular Probes). Further support for the existence of fluid flow from the JGA interstitium through the MD plaque into the tubular lumen was obtained using the in vitro microperfused JGA preparation ( Fig. 6 ). Hydrostatic pressure in the Bowman's capsule was increased by occluding the early proximal tubule with a glass micropipette, a technique which is in principle similar to that of stop flow measurements (oil block) by in vivo micropuncture ( 38 ). The equilibrium pressure in the Bowman's capsule was not measured, but it was most likely less than the constant AA perfusion pressure (50 mmHg). This maneuver resulted in a tremendous increase in the volume of Bowman's space and promptly those of the MD and intercellular lateral spaces ( Fig. 6, A-C ). Significant fluid flow through the MD plaque from the base of the cells toward the tubular lumen was evident in real-time images (video supplement). The morphological basis for the filtration of fluid from the Bowman's capsule back into the JGA interstitium is most likely the presence of Bowman's space and podocyte foot processes in the JGA facing the extraglomerular mesangial interstitium ( 25 ). Fluid flow from the MD base to the tubular lumen is an unexpected finding which conflicts the generally accepted salt reabsorptive model of MD function ( 3 ). However, the majority of the MD transport studies were performed using the in vitro microperfused tubule technique ( 3, 11 ), which may not reflect in vivo conditions. The present studies did not address directionality of salt transport, but there are reports suggesting the possibility of salt secretion in the MD area ( 11, 35 ).

Fluid flow from the JG AA lumen and the Bowman's space into the JGA interstitium appear to be important determinants of the JGA fluid balance and fluid composition. The concentration of MD-derived autacoids is certainly modified by the significant and dynamic JGA fluid flow that was directly visualized in the present study. This supports the earlier hypothesis of Rosivall et al. ( 28, 30 ) on the so-called short-loop feedback mechanism. According to this hypothesis, the TGF and renin release signals (humoral mediators) generated by the classic, long-loop MD feedback are rapidly modulated by the JGA interstitial fluid flow originating from the vascular components of the JGA. However, the present findings are clearly in disagreement with the existing paradigm on the stable JGA environment which is thought to be an ideal milieu for humoral signal transmission ( 2, 32 ). In contrast, the relatively high volume and rapid physiological changes of JGA fluid flow can almost certainly wash MD-derived autacoids (ATP, adenosine, PGE 2, NO, etc.) quickly out of the JGA. Maneuvers that should alter JGA fluid flow, such as acute volume loading, changes in osmolality, and elevations in interstitial pressure indeed modulate TGF sensitivity ( 5, 16, 17 ).

In summary, the present in vivo studies visualized for the first time the filtration of fluid from the fenestrated, renin-positive JG AA segment into the JGA interstitium. Fluid flow in the JGA was significant in volume and rapidly responsive to hemodynamic changes in the JGA. Fluid flow was also observed in the MD area and it appeared to be an important factor in MD cell function. We propose that the functional significance of the JGA fluid flow is twofold: it may help filter the released renin into the renal interstitium (endocrine function) and it may also alter TGF and renin signals in the JGA (hemodynamic function). These studies also suggest that several aspects of the current model of JGA environment and JGA function need to be reevaluated.

GRANTS

These studies were supported by Hungarian Research Grants: OTKA AT 048767, ETT 104/2006, the Hungarian Kidney Foundation, and a Fulbright Fellowship to L. Rosivall and National Institutes of Health Grant DK-064324 to J. Peti-Peterdi.

【参考文献】
  Alund M. Juxtaglomerular cell activity during hemorrhage and ischemia as revealed by quinacrine histofluorescence. Acta Physiol Scand 110: 113-121, 1980.

Barajas L and Latta H. A three-dimensional study of the juxtaglomerular apparatus in the rat. Light and electron microscopic observations. Lab Invest 12: 257-269, 1963.

Bell PD, Lapointe JY, and Peti-Peterdi J. Macula densa cell signaling. Annu Rev Physiol 65: 481-500, 2003.

Bell PD, Lapointe JY, Sabirov R, Hayashi S, Peti-Peterdi J, Manabe K, Kovacs G, and Okada Y. Macula densa cell signaling involves ATP release through a maxi anion channel. Proc Natl Acad Sci USA 100: 4322-4327, 2003.

Boberg U and Persson AE. Tubuloglomerular feedback during elevated renal venous pressure. Am J Physiol Renal Fluid Electrolyte Physiol 249: F524-F531, 1985.

Chu J, Chu S, and Montrose MH. Apical Na + /H + exchange near the base of mouse colonic crypts. Am J Physiol Cell Physiol 283: C358-C372, 2002.

Dunn KW, Sandoval RM, Kelly KJ, Dagher PC, Tanner GA, Atkinson SJ, Bacallao RL, and Molitoris BA. Functional studies of the kidney of living animals using multicolor two-photon microscopy. Am J Physiol Cell Physiol 283: C905-C916, 2002.

Gaillard I, Keramidas M, Liakos P, Vilgrain I, Feige JJ, and Vittet D. ACTH-regulated expression of vascular endothelial growth factor in the adult bovine adrenal cortex: a possible role in the maintenance of the microvasculature. J Cell Physiol 185: 226-234, 2000. <a href="/cgi/external_ref?access_num=10.1002/1097-4652(200011)185:2

Holstein-Rathlou NH and Leyssac PP. TGF-mediated oscillations in proximal intratubular pressure: differences between spontaneously hypertensive rats and Wistar-Kyoto rats. Acta Physiol Scand 126: 333-339, 1986.

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-F502, 2006.

Lapointe JY, Bell PD, and Cardinal J. Direct evidence for apical Na + :2Cl - :K + cotransport in macula densa cells. Am J Physiol Renal Fluid Electrolyte Physiol 258: F1466-F1469, 1990.

Leyssac PP. Changes in single nephron renin release are mediated by tubular fluid flow rate. Kidney Int 30: 332-339, 1986.

Marsh DJ, Sosnovtseva OV, Chon KH, and Holstein-Rathlou NH. Nonlinear interactions in renal blood flow regulation. Am J Physiol Regul Integr Comp Physiol 288: R1143-R1159, 2005.

Molitoris BA and Sandoval RM. Intravital multiphoton microscopy of dynamic renal processes. Am J Physiol Renal Physiol 288: F1084-F1089, 2005.

Moore LC, Iijima K, Rich A, Casellas D, and Goligorsky MS. Communication of the tubuloglomerular feedback signal in the JGA. Kidney Int Suppl 32: S45-S50, 1991.

Navar LG, Inscho EW, Ibarrola M, and Carmines PK. Communication between the macula densa cells and the afferent arteriole. Kidney Int Suppl 32: S78-S82, 1991.

Persson AE, Boberg U, Hahne B, Muller-Suur R, Norlen BJ, and Selen G. Interstitial pressure as a modulator of tubuloglomerular feedback control. Kidney Int Suppl 12: S122-S128, 1982.

Persson BE, Sakai T, and Marsh DJ. Juxtaglomerular interstitial hypertonicity in Amphiuma: tubular origin-TGF signal. Am J Physiol Renal Fluid Electrolyte Physiol 254: F445-F449, 1988.

Peti-Peterdi J, Morishima S, Bell PD, and Okada Y. Two-photon excitation fluorescence imaging of the living juxtaglomerular apparatus. Am J Physiol Renal Physiol 283: F197-F201, 2002.

Peti-Peterdi J, Fintha A, Fuson AL, Tousson A, and Chow RH. Real-time imaging of renin release in vitro. Am J Physiol Renal Physiol 287: F329-F335, 2004.

Peti-Peterdi J. Multiphoton imaging of renal tissues in vitro. Am J Physiol Renal Physiol 288: F1079-F1083, 2005.

Peti-Peterdi J, Komlosi P, Fuson AL, Guan Y, Schneider A, Qi Z, Redha R, Rosivall L, Breyer MD, and Bell PD. Luminal NaCl delivery regulates basolateral PGE 2 release from macula densa cells. J Clin Invest 112: 76-82, 2003.

Peti-Peterdi J. Calcium wave of tubuloglomerular feedback. Am J Physiol Renal Physiol 291: F473-F480, 2006.

Roberts WG and Palade GE. Increased microvascular permeability and endothelial fenestration induced by vascular endothelial growth factor. J Cell Sci 108: 2369-2379, 1995.

Rosivall L and Taugner R. The morphological basis of fluid balance in the interstitium of the juxtaglomerular apparatus. Cell Tissue Res 243: 525-533, 1986.

Rosivall L and Taugner R. Fluid balance in the intersitium of the Goormaghtigh cell field. In: Juxtaglomerular Apparatus and Tubuloglomerular Feedback Mechanism, edited by Persson AEG and Boberg U. Amsterdam: Fernström Foundation Series, Elsevier Science Publisher, vol. 11, 1988, p. 39-50.

Rosivall L, Razga Z, Mirzahosseini S, and Tornoci L. Endothelial permeability of the afferent arteriole and its changes as the result of alteration in the activity of the renin-angiotensin system. J Am Soc Nephrol Suppl 11: S172-S177, 1999.

Rosivall L. Morphology and function of the distal part of the afferent arteriole. Kidney Int Suppl 30: S10-S15, 1990.

Rosivall L, Razga Z, and Ormos J. Morphological characterization of human juxtaglomerular apparatus. Kidney Int Suppl 32: S9-S12, 1991.

Rosivall L and Peti-Peterdi J. Heterogeneity of the afferent arteriole: correlations between morphology and function. Nephrol Dial Transpl 21: 2703-2707, 2006.

Salmon AH, Neal CR, Bates DO, and Harper SJ. Vascular endothelial growth factor increases the ultrafiltration coefficient in isolated intact Wistar rat glomeruli. J Physiol 570: 141-156, 2006.

Schnermann J, Weihprecht H, Lorenz JN, and Briggs JP. The afferent arteriole-the target for macula densa-generated signals. Kidney Int Suppl 32: S74-S77, 1991.

Schnermann J and Levine DZ. Paracrine factors in tubuloglomerular feedback: adenosine, ATP, and nitric oxide. Annu Rev Physiol 65: 501-529, 2003.

Schnermann J and Briggs J. Function of the juxtaglomerular apparatus: local control of glomerular hemodynamics. In: The Kidney, edited by Seldin DW and Giebisch G. New York: Raven, 1985, p. 669-697.

Schnermann J, Briggs J, and Schubert G. In situ studies of the distal convoluted tubule in the rat. I. Evidence for NaCl secretion. Am J Physiol Renal Fluid Electrolyte Physiol 243: F160-F166, 1982.

Sorensson J, Fierlbeck W, Heider T, Schwarz K, Park DS, Mundel P, Lisanti M, and Ballermann BJ. Glomerular endothelial fenestrae in vivo are not formed from caveolae. J Am Soc Nephrol 13: 2639-2647, 2002.

Taugner R, Rosivall L, Buhrle CP, and Groschel-Stewart U. Myosin content and vasoconstrictive ability of the proximal and distal (renin-positive) segments of the preglomerular arteriole. Cell Tissue Res 248: 579-588, 1987.

Wright FS and Schnermann J. Interference with feedback control of glomerular filtration rate by furosemide, triflocin, and cyanide. J Clin Invest 53: 1695-1708, 1974.

Yu W, Sandoval RM, and Molitoris BA. Quantitative intravital microscopy using a Generalized Polarity concept for kidney studies. Am J Physiol Cell Physiol 289: C1197-C1208, 2005.

作者单位：1 Hungarian Academy of Sciences and Semmmelweis University Nephrology Research Group, Institute of Pathophysiology, Semmelweis University Faculty of Medicine, Budapest, Hungary; and 2 Departments of Physiology and Biophysics and Medicine, Zilkha Neurogenetic Institute, University of Southern Califor