点击显示 收起
【摘要】 This study was performed to evaluate the alterations of glomerular filtration barrier characteristics following acute renal ischemia-reperfusion (I/R). Ischemia was induced in anesthetized rats by unilateral renal artery occlusion for either 20 or 60 min, followed by reperfusion during 20 or 60 min, respectively, with the contralateral kidney serving as control. Sieving coefficients ( ) were obtained by analyzing Ficoll [mol.radius ( a e ) 13-85 Å] in urine and plasma after 20 and 60 min I/R. Furthermore, for human serum albumin (HSA) was estimated using a tissue uptake technique after 20 and 60 min of I/R, while clearance of HSA compared with that for neutralized HSA (nHSA) was assessed after 20 min of I/R only. Glomerular filtration rate (GFR) was measured by [ 51 Cr]EDTA and inulin. I/R reduced GFR and increased for Ficoll molecules of a e 55 Å and for albumin. for Ficoll vs. a e, analysed using a two-pore model, demonstrated that, despite increases in, the large-pore fractional ultrafiltration coefficient ( L ) was unchanged after 20 min of I/R, owing to the decline in GFR, but increased after 60 min of I/R. However, the apparent L for albumin increased already after 20 min of I/R ( P < 0.005) and the nHSA/HSA clearance ratio was slightly reduced, possibly reflecting a diminished negative charge barrier. In conclusion, after 20 min of I/R, indications of a reduced charge selectivity were noted, while after 60 min of I/R, there was mainly a reduction in size selectivity, compatible with an increased formation of large pores.
【关键词】 albumin Ficoll anoxia fractional clearance oxidative stress
ISCHEMIA REPRESENTS THE MOST common cause of acute renal failure (ARF). When renal blood flow is compromised, oxygen is depleted from the tissue resulting in a diminution of ATP. This reduces the capacity of the Na-K-ATPase, producing cell swelling. During reperfusion, free oxygen radicals are formed, further contributing to cellular damage. The effects of ischemia-reperfusion (I/R) on tubular epithelial cell damage have been studied extensively, particularly the decreased capacity for tubular protein reuptake, leading to proteinuria. Following I/R, the function of the glomerular barrier is clearly also affected. However, this has only rarely been addressed previously.
The glomerular barrier is made up by three sequential layers: the fenestrated endothelium with its glycocalyx, the glomerular basement membrane (GBM), and finally, the podocytes with their interdigitating foot processes and the podocyte slit diaphragms (PSD) ( 6 ). All of these three layers are potential ischemic targets, and alterations in any of them would lead to an increased glomerular permeability. The first layer, the endothelial glycocalyx, is composed of negatively charged proteoglycans that, conceivably, contribute to the negative charge of the glomerular barrier. Morphological studies in heart capillaries have indicated that ischemia causes disruption and "clumping" of the endothelial glycocalyx ( 16, 30 ), hence suggesting that the disruption of glomerular glycocalyx may play a role during ischemia-induced proteinuria. However, the endothelial glycocalyx is not visualized by normal fixation procedures and is therefore overlooked in many pathological situations. Since ischemia does not cause any obvious ultrastructural changes of the glomerular filter, it has been proposed that the proteinuria-induced by ischemia may, at least partly, be a result of injury to the glycocalyx.
The second layer, the GBM, is composed of collagen type IV, laminin-11, entactin and heparan sulphate proteoglycans (HSPGs), such as perlecan and agrin ( 9 ). HSPGs are important components of the GBM, and their negative charge may be of both structural and functional importance. This is evident in animals overexpressing heparanase-1, an enzyme degrading HSPG chains, the ensuing alteration in the GBM resulting in proteinuria ( 33 ). Furthermore, increased concentrations of metalloproteinases have been detected in glomeruli after I/R ( 4 ) and may be a factor contributing to ischemic proteinuria, as these enzymes are important for normal degradation and remodelling of the GBM.
It has been shown that podocytes undergo flattening and spreading of their foot processes in ARF ( 20 ). These changes occur early, already during the ischemic period, and increase in severity with the ischemic duration. There is a close structural interaction between the GBM and the podocytes. In mice with altered GBM, the podocytes appear effaced with a reduced number of foot processes ( 33 ). Hence, following I/R, alterations clearly occur in all layers of the glomerular capillary wall.
The purpose of the present study was to investigate the functional nature of the lesions occurring in the glomerular capillary wall following acute I/R injury. To determine whether acute I/R leads to an alteration in the size-selective barrier, we investigated the urine excretion of FITC-Ficoll, a neutral polydisperse polysaccharide, allowing the assessment of sieving coefficients (fractional clearances, ) for a broad spectrum of molecular sizes in one single experiment. To get further insights into the potential charge-dependent nature of permeability changes, we also assessed the fractional clearance of native (neg.) albumin. Moreover, when permeability changes were very discrete, i.e., after 20 min I/R, we compared the clearance of native (neg. charged) albumin to that of neutralized albumin.
EXPERIMENTAL PROCEDURES
Experiments were performed in male Wistar rats (Møllegaard, Lille Stensved, Denmark). The experimental groups are presented in Table 1. In addition, one group of animals ( n = 7) were used for assessing the clearance ratio of neutralized albumin to native (neg.) albumin after 20 min of I/R. The rats were kept on standard chow and were allowed free access to food and water until the day of the experiments. The experiments were approved by the Animal Ethics Committee at Lund University and conducted according to the American Physiological Society Guiding Principles in the Care and Use of Laboratory Animals.
Table 1. GFR and body weight for the different experimental groups
Surgery. Anesthesia was induced with pentobarbital sodium (60 mg/kg ip) and body temperature was kept at 37°C by a thermostatically controlled heating pad. The tail artery was used for continuous monitoring of arterial pressure on a polygraph (Grass instruments model 7B, Quincy, MA). A tracheotomy was performed to facilitate breathing. The left carotid artery and the left jugular vein were cannulated (PE-50) for blood sampling and infusion purposes, respectively. Access to the abdominal cavity was obtained through a midline abdominal incision. The urethers were dissected free and cannulated (PE-10) for continuous urine sampling. Furosemide (Impugan, Dumex-Alpharma, 0.5 mg) was administered in the tail artery to increase urine production and facilitate cannulation of the ureters.
Experimental protocols. Glomerular filtration rate (GFR) was measured in both kidneys separately during the preischemic resting period (20-30 min), using 51 Cr-EDTA. A priming dose of 51 Cr-EDTA (100 µl, 0.37 MBq, Amersham Biosciences, Buckinghamshire, UK) was administered and followed by a continuous infusion (3 ml/h) of 51 CrEDTA (0.37 MBq/ml in 0.9% NaCl) throughout the experiment. Ischemia of the left kidney was induced with a temporary 20- or 60-min clamp of the left renal artery. Urine was collected from the urethers (no urine was produced from the clamped kidney during the artery obstruction). After the ischemic period (20 or 60 min), the artery clamp was removed, followed by a 20- or 60-min reperfusion period, respectively. During the reperfusion period, urine production resumed in the previously obstructed kidney. Ten minutes before the end of the reperfusion period, a bolus dose containing FITC-labeled Ficoll-70 (42 µg), FITC-Ficoll-400 (1 mg), and FITC-inulin (0.5 µg; TdB Consultancy, Uppsala, Sweden) was given followed by an infusion of FITC-Ficoll-70 (94.5 µg/min), FITC-Ficoll-400 (3 mg/min), FITC-inulin (1.5 µg/min), and 51 Cr-EDTA ( n = 7 at 20 min and n = 9 at 60 min). At the end of the reperfusion period, urine was collected for 5 min and a midpoint plasma sample was aspirated for analysis of Ficoll sieving coefficients.
Tissue uptake technique. After 20 or 60 min of I/R, albumin was measured using a tissue uptake technique ( 17 ). [ 125 I]human serum albumin (HSA) was given as a bolus (0.2 MBq, Institute for Energy Technique, Kjeller, Horten, Norway) in the tail artery ( n = 11 at 20 min and n = 6 at 60 min). Six blood samples (25 µl) and one urine sample were collected during 8 min. Thereafter, a whole body washout was performed via the carotid artery (20 ml/min) for 8 min. The inferior vena cava was freed and cut open for collection of the rinse fluid. The washout fluid mixture contained equal amounts of 0.9% saline and heparinized horse serum (SVA, Uppsala, Sweden). Thereafter, the kidneys were removed and the cortex was dissected and assessed with respect to radioactivity. Sieving coefficients for neutralized human serum albumin (nHSA) were measured after 20 min of I/R ( n = 6) as described above for [ 125 I]HSA. In seven additional experiments, we simultaneously compared the clearance of [ 125 I]HSA to that of [ 131 I]nHSA after 20-min ischemia. The simultaneous measurement of [ 125 I]HSA and [ 131 I]nHSA did not allow the assessment of GFR by [ 51 Cr]EDTA. Hence, only a clearance ratio, but not absolute values of, could be obtained in these experiments. nHSA was prepared by Dr. Olav Tenstad by a graded modification of the COOH groups using a procedure modified from Hoare and Koshland ( 11 ) and earlier described at some length ( 17 ). nHSA was labeled with 131 I, using 1,3,4,6-tetrachloro-3,6 -diphenylglycouril (Iodo-Gen) as described in ( 17 ). All radioactivity measurements were performed in a gamma scintillation counter (Wizard 1480, LKP Wallac, Turku, Finland). Radioactive decay and spillover from the 51 Cr to the 125 I channel, or from the 131 I channel to the 125 I channel, were appropriately accounted for.
High-performance size exclusion chromatography. Size exclusion was achieved by using an Ultrahydrogel-500 column (Waters) and a phosphate buffer with 0.15 M NaCl (pH 7.4). Fluorescence was detected with a fluorescence detector with a exitation at 492 nm and a emission at 518 nm (Waters 2475). The system was controlled by Breeze software 3.2 (Waters). The column was calibrated using five narrow FITC-Ficoll standards, seven narrow FITC-dextran standards and a few protein standards. A calibration curve was achieved using the relationship: y = -475.45 x 3 + 1,059.4 x 2 - 878.78 x + 294.41 and is described elsewhere ( 1 ).
Calculations. For the tissue uptake technique, renal tracer protein clearance was calculated from the amount of tracer radioactivity accumulated in both kidneys (cortex) plus the TCA-precipitable urine tracer activity (collected during the tracer infusion period) divided by the area under the curve of the plasma tracer concentration vs. time function. Protein values were calculated by dividing the measured protein clearance by the simultaneously assessed GFR ( 17, 27 ). Ficoll were obtained by analyzing the HPSEC-curves obtained from the urine and the plasma sample (Cp F ) for each experiment. The urine concentration vs. a e curve was divided by the inulin concentration, to obtain the primary urine concentrations (Cu F ). for each a e was calculated by dividing Cu F by Cp F (Cu F /Cp F ). The two-pore model ( 17, 22, 27 ) was used to analyze the data for Ficoll (15-80Å). A nonlinear least squares regression analysis was used to obtain the best curve fit, using scaling multipliers as described previously ( 17 ). An apparent L was calculated for the HSA data, as obtained from the following equation [see Eq. 4 in ( 27 ) and Eq. 25 in ( 22 )]:
Because (1 - L ) is very close to unity we have:
where was set at 28 mmHg, and hydraulic conductance (L p S) was calculated assuming a filtration pressure of 9 mmHg.
Statistical analysis. Values are presented as means ± SE. Differences between groups were calculated using one-way ANOVA with Bonferroni's multiple-comparison correction. A paired t -test was used for comparisons between HSA and nHSA in the same kidney. Significance levels were set at * P < 0.05, ** P < 0.001, and *** P < 0.005.
RESULTS
20 min of I/R. While GFR remained unchanged in the control kidney during both the ischemic and reperfusion period, GFR was markedly reduced ( 45%) after the reperfusion period in the ischemic kidney ( Table 1 ). From the relationship of log for Ficoll vs. SE- radius (15-80 Å), it is evident that for FITC-Ficoll increased for 55Å in the postischemic kidney (2.7-fold) compared with the control kidney, as shown in Fig. 1. The best curve fit of vs. a e according to the two-pore theory was obtained for the parameters listed in Table 3. Even though was significantly higher for large size Ficoll (a e 55Å) after 20 min of I/R, the fractional ultrafiltration coefficient accounted for by the large pores ( L ) was not significantly increased, indicating that the reduced GFR postischemia may have been the prime factor responsible for the increase in Ficoll ( Fig. 2 ) (see Ref. 23 ).
Fig. 1. Sieving coefficients ( ) vs. SE-radius (a e ) for Ficoll and HSA in the control kidney and in the ischemic kidney at 20 and 60 min of ischemia-reperfusion (I/R). Control (20 min) is shown in red, control (60 min) in blue, ischemia (20 min) in purple, and ischemia (60 min) in green. Ficoll are given for 400 data points between 15 and 80 Å and for HSA is given at 36 Å.
Table 3. Two-pore analysis
Fig. 2. Relative fractional ultrafiltration coefficient accounted for by large pores ( L ) in percent for Ficoll and native albumin (HSA). L for Ficoll was obtained from the 2-pore model. L for HSA was calculated assuming a filtration pressure of 9 mmHg (to assess the LpS) and was set at 28 mmHg while the large pore radius was assumed to be 100 Å as obtained from the Ficoll data. Absolute values for L are shown in Tables 2 and 3.
The for neutralized HSA was ninefold higher than that for native HSA in the control kidney ( P < 0.005; Table 2 ). Both HSA and nHSA increased after the I/R insult, the increase however not reaching statistical difference for nHSA ( Table 2 ). In the control kidney 99 ± 0.083% of the nHSA and 88 ± 2.8% of the native HSA was reabsorbed in the proximal tubules, while in the ischemic kidney the fractional reabsorption was 97 ± 0.49% and 81 ± 3.4%, respectively.
Table 2. Sieving coefficients ( ) for HSA and nHSA, and apparent L for HSA at 20 and 60 min of I/R, respectively
In a separate set of experiments, in which native (neg.) and neutral albumin were analyzed together simultaneously (control kidney vs. ischemic kidney), the ratio of clearance for nHSA to HSA was slightly reduced ( P = 0.049) in the kidneys exposed to 20 min of I/R compared with control ( Fig. 3 ), indicative of a reduction in the negative charge barrier. This was further evidenced by the calculation of an "apparent" L for HSA according to the two pore theory (assuming a large pore radius of 100 Å). After 20 min I/R, there was an increase in this parameter ( Table 2 and Fig. 2 ). An increase in apparent L for native HSA, but not for the uncharged Ficoll, may thus reflect a change in the glomerular charge barrier characteristics ( Fig. 3 ). The small pore radius (r s ) and large pore radius (r L ) were unchanged after the I/R insult ( Table 3 ).
Fig. 3. The clearance of nHSA and HSA were determined simultaneously using the tissue uptake technique in a separate series of experiments ( n = 7) to assess the nHSA/HSA ratio in the ischemic and nonischemic kidneys. The reduction in nHSA/HSA ratio in the ischemic kidney after 20 min of I/R was of borderline statistical significance ( P = 0.049).
60 min of I/R. Ischemia and reperfusion of the contralateral kidney did not affect hemodynamics (GFR) in the ipsilateral control kidney. Sixty minutes of I/R, however, markedly reduced the GFR in the postischemic kidney ( 75%; Table 1 ). The for Ficoll molecules of a e 55Å increased 5-fold in the I/R kidney, the increase being 2.4-fold larger than for 20 min I/R ( Fig. 1 ). The r s and r L were not affected by 60 min I/R. However, there was significant increases in L, indicating that 60 min of I/R increased the number of large pores in the glomerular filter ( Fig. 2 and Table 3 ), without affecting the large pore radius. A 0 / X was reduced threefold ( P < 0.005) after 60 min of I/R, i.e., largely in parallel with the fall in GFR ( Table 3 ), indicating a reduced effective pore area for solute diffusion after 60 min of I/R.
DISCUSSION
Despite extensive experimental and clinical research on ARF, only modest progression have been made to improve the mortality during the last 20 years. ARF is common in critically ill patients, the most frequent cause being an ischemic injury to the kidney, initiated by, for example, congestive heart failure, hypotension, septic shock, volume depletion, the use of aminoglycosides or radiocontrast procedures ( 25 ). There exist a spectrum of treatments for ARF, but they are all currently only supportive. Commonly practiced treatments that are used to inhibit development of ARF include: volume expansion, loop-diuretics and mannitol, dopamine, N -acetylcysteine, and atrial natriuretic peptide (ANP). However, recent clinical studies show no clear therapeutic benefit of any of the pharmacologic therapies for ARF ( 19 ), and further research is needed to increase the knowledge in this field.
The effects of I/R on glomerular barrier selectivity have only been rarely investigated. Hence, this is the first study, to our knowledge, in which a detailed analysis has been performed to evaluate the nature of the glomerular permeability changes occurring after acute I/R in vivo. The major result of the present study is that acute I/R markedly reduces the size selectivity of the glomerular barrier by the functional appearance of an increased number of large pores (cf. "shunt pathways"). This pattern of alterations of the glomerular barrier function mimics what has been previously seen in a number of different pathophysiologic disorders. The increased number of large pores in the present study after 60 min of I/R, as determined by Ficoll, was mirrored by a large increase in for albumin (a e = 36 Å; Fig. 1 ). According to the two-pore theory, native albumin, due to its net negative charge, is for its transglomerular passage more or less totally confined to the large pore pathway. This is because of the high size and charge selectivity of the glomerular barrier to proteins ( 21, 23, 27, 29 ), normally excluding (neg.) native albumin from the small pore pathway. An increase in the large pore number (mirrored by L ) should thus be well reflected by increases in for albumin, as noted in the present study.
Just a minor (nonsignificant) increase in the fractional ultrafiltration coefficient accounted for by the large pores ( L ) was observed after 20 min of I/R for the Ficoll sieving data. By contrast, L for Ficoll increased significantly (4-fold) after 60 min of I/R ( P < 0.05; Fig. 2 ). Assessed from for native albumin, however, an increase in apparent L was observed already at 20 min of I/R, but no further increase was observed after 60 min I/R. Since Ficoll data predicted only a minor, non-significant increase in large pore number after 20 min of I/R, the increase in apparent L for HSA may be due to a small charge defect in the glomerular barrier appearing after 20 min of I/R, making albumin partly permeable also through the small pore pathway. This was further corroborated by the fact that the clearance ratio of nHSA to HSA was slightly reduced at 20 min I/R ( Fig. 3 ). However, after 60 min of I/R, the increase in large pore number, as interpreted from the increase in L for Ficoll, apparently superseded the small charge defect observed at 20 min I/R, because L for albumin (in absolute and relative terms) did not differ from that calculated from Ficoll data in this situation.
There were no changes in Ficoll among small Ficoll molecules (a e <50 Å) after the I/R injury. The reduction in GFR following upon I/R would per se lead to an increase in for these Ficoll molecules provided that the capillary pore area (A 0 / x) and small pore radius (r s ) had remained unchanged, leading to a "steeper" sieving curve ( 23 ). However, this effect was canceled by the simultaneous parallel reduction in A 0 / x, thus leaving the Peclet numbers (ratio of GFR to A 0 / x), and hence, the sieving coefficients in the 20-50Å a e -interval, unchanged. The parallel decrease in A 0 / x and GFR could be attributed to alterations in renal hemodynamics. After ischemia, swelling of endothelial cells and a reduction of the capillary lumen have been reported. The ensuing decline in glomerular blood flow has been denoted the "no-reflow" phenomenon ( 26 ). There are also indications of a reduced NO production in the ischemic endothelium, related to the formation of ROS ( 1 ), also contributing to the "no-reflow" phenomenon. Indeed, intravital videomicroscopy of glomerular vessels during reperfusion of a previously obstructed kidney has demonstrated an instantaneous recovery of glomerular blood flow, followed by an oscillating flow pattern of cessation and partial recovery of glomerular perfusion ( 31 ).
Although the increment in for native (neg. charged) albumin (a e =36 Å) in the present study largely paralleled the increase in for large MW (a e 55 Å) Ficoll molecules, for Ficoll of 36Å radius (Ficoll 36Å ), remained largely unchanged. This may be explained by the fact that Ficoll 36Å was two orders of magnitude more permeable than native albumin and one order of magnitude more permeable than neutralized albumin of equivalent molecular radius across the glomerulus, as discussed at some length previously ( 23, 29 ). The relative "hyperpermeability" of Ficoll 36Å will make this molecule relatively insensitive to changes in large pore number and large pore volume flow ( 23, 29 ). In the present study 60 min of I/R caused the large pore fractional fluid flow ( J vL /GFR) to increase from 3.6·10 -4 to 16·10 -4, i.e., by 0.0012. Such changes would theoretically alter for Ficoll 36Å insignificantly, i.e., by only 0.5% (from 0.10 to 0.1005), assuming an unchanged small and large pore radius. At the same time, one would observe a twofold increase in for HSA (from 0.0005 to 0.001), provided that no charge alterations had occurred and that albumin transport is confined to large pores. Again, the fact that alb increased more than twofold may be ascribed to slight reductions in microvascular charge barrier.
Although Ficoll seems to overestimate the clearance of proteins in the a e interval of 25-55 Å, there is a remarkable similarity in sieving coefficients for proteins and Ficoll in the a e interval 55-75Å. As discussed in a previous paper, this probably is due to the fact that the a e /r s ratio for this molecular size range is approximately <0.5-0.7, in which the anomalous glomerular permeation behavior of Ficoll is not manifested ( 2, 23 ). Because both proteins and Ficoll 55Å can be used to assess the large pore radius and L, we have tried to reconcile albumin with those for Ficoll 55-75Å. This analysis yielded a slightly higher value for L and a slightly lower value for the large pore radius than that obtained from Ficoll data alone ( Fig. 4 ). It thus seems that for albumin may have been (slightly) higher than expected from just extrapolating Ficoll 55-75Å data back to 36 Å. The reason for this discrepancy is not obvious. It may be due to a slight overestimation of for albumin, especially in the ischemic situation, partly due to postischemic interstitial trapping of tracer that may not be easily washed out ( 7 ). could also be slightly overestimated due to trapping of proteins in the endothelium and in the mesangium. However, we believe that the fraction of trapped proteins/polysaccharides is, after all, very small since our data comply with a recent, very carefully performed micropuncture study ( 28 ). Alternatively, the large pore radius was slightly overestimated from Ficoll data alone. One reason for such an overestimation could be the predicted Ficoll hyperpermeability, for very large Ficoll molecules, i.e., when a e /r L 0.7 (for a e 70-75Å).
Fig. 4. Both native albumin and Ficoll 55Å are, according to the 2-pore theory, confined for their glomerular passage to the large pore pathway because neg. albumin cannot pass 38Å (charged) small pores. Reconciling for albumin with for Ficoll 55-75Å and fitting the data to a single large pore yielded a slightly higher value for L (control 20 : 3.3 x 10 -4, ischemia 20 : 6.3 x 10 -4, control 60 : 2.9 x 10 -4 ischemia 60 : 18 x 10 -4 ) and at the same time a slightly decreased large pore radius ( 92 Å compared with 100 Å), than what was obtained from Ficoll data alone ( Table 3 ). Solid lines represent the best fit of control groups, dotted line represents 20 min of I/R, and dashed line represents 60 min of I/R. Data points for control 20 are shown as, ischemia 20 as, control 60 as, and ischemia 60 as open pentagons.
The major rationale for using polysaccharides as molecular markers for glomerular transport is that their tubular reabsorption is negligible ( 24 ) so that the final urine can be used to investigate the sieving properties of the glomerular filter. During ischemic conditions, the tubular epithelial cells become damaged and the tubular reabsorption decreases, which will influence the reabsorption, primarily of proteins. In the present study, 88% of the HSA and 99% of the nHSA was reabsorbed in the control kidney, whereas only 81% of the HSA and 97% of the nHSA was reabsorbed in the ischemic kidney (20 min of I/R). The larger amount of tubular reabsorption of the neutralized HSA is in agreement with earlier studies, showing that the higher the isoelectric point of a protein, the larger is its proximal tubular reuptake ( 5 ). However, since both the urine excretion and the renal epithelial uptake of protein tracer was used to calculate the clearance and in our study, the measured was insensitive to the differences in proximal tubular reabsorption among the different tracers.
The endothelial glycocalyx is thought to represent the major charge barrier in renal and systemic capillaries ( 10 ). Modifying the glycocalyx by enzymatic digestion using hyaluronidase, increased the glomerular permeability to both albumin and large size Ficoll molecules in vivo ( 12 ). Furthermore, treatment with another glycocalyx-degrading enzyme, chondroitinase, in the cooled isolated perfused kidney (cIPK), apparently reduced the thickness of the glycocalyx while increasing the permeability ( ) to albumin. However, in those experiments the increases in for large size Ficolls were not significant ( 13 ). The results mimic the changes observed in the present study after 20 min of I/R (mild ischemia), suggesting that this relatively short I/R period may have led to alterations that could, at least partly, be attributed to alterations in the endothelial glycocalyx. The reactive oxygen species (ROS) formed during the reperfusion period ( 8 ) can alter molecular structures and induce an inflammatory response. According to a previous study infusion of H 2 O 2 for a period of 1 h, indeed induced an increase in urine albumin excretion and in for large 42 Å) dextrans, without causing significant changes in GFR, again indicating the induction of an increased number of large pores in the glomerular filter ( 32 ). Although the authors did not find any apparent alterations in the anionic barrier of the GBM, assessed by staining with polyethyleneimine, there could still have been marked alterations in the endothelial glycocalyx in addition to the reduced size selectivity.
The pathophysiology of the permeability changes occurring after severe ischemia (60 min of I/R) is not exactly known, but could be ascribed to a number of different mechanisms. One scenario could be a remodelling of the GBM. In isolated renal endothelial cells from rats which had undergone 30 min of ischemia, there was an increase in the proteolytic activity of matrix metalloproteinases (MMPs) ( 3 ), enzymes responsible for matrix (collagen) remodelling. Furthermore, after renal I/R both mRNA and protein levels of MMP-9 increased ( 4 ), indicating that renal ischemia could lead to degradation of the GBM, which in turn could be responsible for the increase in glomerular permeability. HSPGs are important structural contributors in the GBM, and gene modifications leading to either loss of HSPGs, or their degradation, indeed leads to proteinuria ( 18, 33 ). Thus both HSPGs and collagen may be targets of ischemic damage.
Ischemia has been shown to cause flattening and spreading of the podocyte foot processes, a condition that occurs early in the ischemic response ( 20 ). There are tight connections between the GBM and the podocytes. The architecture of the foot processes is dependent on integrin- ( 3 1) and -dystroglycan-mediated interactions with the GBM, -dystroglycan being linked to laminin and agrin. Recently, it has been shown that these linkages could be disrupted by ROS in in vitro experiments ( 14 ). Hence, the increased amount of ROS associated with ischemia could, in part, explain the detachment of the podocyte processes from the GBM. Whether the detachment is a primary phenomenon or a secondary consequence occurring due to remodelling of the GBM is not clear. In mice overexpressing heparanase-1, the podocytes appear flattened and have a reduced number of foot processes ( 33 ), probably secondary to the degradation of the GBM. Likewise, mice lacking the 3-integrin are born with disorganized GBM and nondifferentiated podocytes lacking foot-processes ( 15 ). Thus it seems that any condition in which GBM and the podocytes loose their interconnections may lead to alterations in the glomerular size selective barrier.
In conclusion, 60 min of I/R induced an increased formation of large pores in the glomerular filter, reflected by increases in L for both Ficoll and native albumin. However, after 20 min of I/R, the insult to the glomerular filter was less pronounced and compatible with reductions in charge selectivity, and to a lesser extent, in size selectivity, reflected by a larger increment in the glomerular permeability to albumin than to Ficoll.
GRANTS
This study was supported by Swedish Medical Research Council Grant 08285, the Lundberg Medical Foundation, European Union Contract FMRX-C98-2019, and Lars Hierta Memorial Foundation.
ACKNOWLEDGMENTS
Dr. Venturoli is greatly appreciated for contributing with mathematical advice and fruitful discussions.
【参考文献】
Asgeirsson D, Venturoli D, Rippe B, and Rippe C. Increased glomerular permeability to negatively charged Ficoll relative to neutral Ficoll in rats. Am J Physiol Renal Physiol. First published May 30, 2006; doi:101152/ajprenal.00488.2005.
Beresewicz A, Czarnowska E, and Maczewski M. Ischemic preconditioning and superoxide dismutase protect against endothelial dysfunction and endothelium glycocalyx disruption in the postischemic guinea-pig hearts. Mol Cell Biochem 186: 87-97, 1998.
Bohrer MP, Patterson GD, and Carroll PJ. Hindered diffusion of dextran and Ficoll in microporous membranes. Macromolecules 17: 1170-1173, 1984.
Caron A, Desrosiers RR, and Beliveau R. Ischemia injury alters endothelial cell properties of kidney cortex: stimulation of MMP-9. Exp Cell Res 310: 105-116, 2005.
Caron A, Desrosiers RR, Langlois S, and Beliveau R. Ischemia-reperfusion injury stimulates gelatinase expression and activity in kidney glomeruli. Can J Physiol Pharmacol 83: 287-300, 2005.
Christensen EI, Rennke HG, and Carone FA. Renal tubular uptake of protein: effect of molecular charge. Am J Physiol Renal Fluid Electrolyte Physiol 244: F436-F441, 1983.
Deen WM, Lazzara MJ, and Myers BD. Structural determinants of glomerular permeability. Am J Physiol Renal Physiol 281: F579-F596, 2001.
Folmerz P, Hansson R, Jonsson O, Lundstam S, Pettersson S, Rippe B, and Schersten T. Effects of ischaemia on leakage of albumin in rabbit kidneys. Acta Physiol Scand 131: 103-107, 1987.
Granger DN and Korthuis RJ. Physiologic mechanisms of postischemic tissue injury. Annu Rev Physiol 57: 311-332, 1995.
Groffen AJ, Ruegg MA, Dijkman H, van de Velden TJ, Buskens CA, van den Born J, Assmann KJ, Monnens LA, Veerkamp JH, and van den Heuvel LP. Agrin is a major heparan sulfate proteoglycan in the human glomerular basement membrane. J Histochem Cytochem 46: 19-27, 1998.
Haraldsson B and Sörensson J. Why do we not all have proteinuria? An update of our current understanding of the glomerular barrier. News Physiol Sci 19: 7-10, 2004.
Hoare DG and Koshland DE Jr. A method for the quantitative modification and estimation of carboxylic acid groups in proteins. J Biol Chem 242: 2447-2453, 1967.
Jeansson M and Haraldsson B. Glomerular size and charge selectivity in the mouse after exposure to glucosaminoglycan-degrading enzymes. J Am Soc Nephrol 14: 1756-1765, 2003.
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-F116, 2006.
Kojima K, Davidovits A, Poczewski H, Langer B, Uchida S, Nagy-Bojarski K, Hovorka A, Sedivy R, and Kerjaschki D. Podocyte flattening and disorder of glomerular basement membrane are associated with splitting of dystroglycan-matrix interaction. J Am Soc Nephrol 15: 2079-2089, 2004.
Kreidberg JA, Donovan MJ, Goldstein SL, Rennke H, Shepherd K, Jones RC, and Jaenisch R. Alpha 3 beta 1 integrin has a crucial role in kidney and lung organogenesis. Development 122: 3537-3547, 1996.
Lindner JR, Ismail S, Spotnitz WD, Skyba DM, Jayaweera AR, and Kaul S. Albumin microbubble persistence during myocardial contrast echocardiography is associated with microvascular endothelial glycocalyx damage. Circulation 98: 2187-2194, 1998.
Lund U, Rippe A, Venturoli D, Tenstad O, Grubb A, and Rippe B. Glomerular filtration rate dependence of sieving of albumin and some neutral proteins in rat kidneys. Am J Physiol Renal Physiol 284: F1226-F1234, 2003.
Morita H, Yoshimura A, Inui K, Ideura T, Watanabe H, Wang L, Soininen R, and Tryggvason K. Heparan sulfate of perlecan is involved in glomerular filtration. J Am Soc Nephrol 16: 1703-1710, 2005.
Palevsky P. Acute renal failure. NephSAP 2: 41-76, 2003.
Racusen LC, Prozialeck DH, and Solez K. Glomerular epithelial cell changes after ischemia or dehydration. Possible role of angiotensin II. Am J Pathol 114: 157-163, 1984.
Renkin EM and Gilmore JP. Glomerular filtration. In: Handbook of Physiology. Renal Physiology, edited by Orloff J and Berliner RW. Washington, D.C.: Am Physiol Soc, 1973, p. 185-248.
Rippe B and Haraldsson B. Transport of macromolecules across microvascular walls. The two-pore theory. Physiol Rev 74: 163-219, 1994.
Rippe C, Asgeirsson D, Venturoli D, Rippe A, and Rippe B. Effects of glomerular filtration rate on Ficoll sieving coefficients (theta) in rats. Kidney Int In press.
Schaeffer RC Jr, Gratrix ML, Mucha DR, and Carbajal JM. The rat glomerular filtration barrier does not show negative charge selectivity. Microcirculation 9: 329-342, 2002.
Star RA. Treatment of acute renal failure. Kidney Int 54: 1817-1831, 1998.
Summers WK and Jamison RL. The no reflow phenomenon in renal ischemia. Lab Invest 25: 635-643, 1971.
Tencer J, Frick IM, Öqvist BW, Alm P, and Rippe B. Size-selectivity of the glomerular barrier to high molecular weight proteins: upper size limitations of shunt pathways. Kidney Int 53: 709-715, 1998.
Tojo A and Endou H. Intrarenal handling of proteins in rats using fractional micropuncture technique. Am J Physiol Renal Fluid Electrolyte Physiol 263: F601-F606, 1992.
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-F613, 2005.
Ward BJ and Donnelly JL. Hypoxia induced disruption of the cardiac endothelial glycocalyx: implications for capillary permeability. Cardiovasc Res 27: 384-389, 1993.
Yamamoto T, Tada T, Brodsky SV, Tanaka H, Noiri E, Kajiya F, and Goligorsky MS. Intravital videomicroscopy of peritubular capillaries in renal ischemia. Am J Physiol Renal Physiol 282: F1150-F1155, 2002.
Yoshioka T, Ichikawa I, and Fogo A. Reactive oxygen metabolites cause massive, reversible proteinuria and glomerular sieving defect without apparent ultrastructural abnormality. J Am Soc Nephrol 2: 902-912, 1991.
Zcharia E, Metzger S, Chajek-Shaul T, Aingorn H, Elkin M, Friedmann Y, Weinstein T, Li JP, Lindahl U, and Vlodavsky I. Transgenic expression of mammalian heparanase uncovers physiological functions of heparan sulfate in tissue morphogenesis, vascularization, and feeding behavior. FASEB J 18: 252-263, 2004.
作者单位：Department of Nephrology, University Hospital of Lund, Lund, Sweden