| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Departments of Pathology and Laboratory
Medicine*
and Medicine,
University of North Carolina at Chapel Hill, Chapel Hill, North
Carolina; the Departments of Clinical
Immunology
and
Pathology,§
University Hospital,
Groningen, The Netherlands; and the Department of
Pathology,¶
University Hospital, Nijmegen,
The Netherlands
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Increased NO production has been demonstrated in several experimental models of glomerulonephritis, most likely derived from infiltrating inflammatory cells expressing iNOS.6,7 Increased production of NO radicals can be cytotoxic and cytostatic and may contribute to tissue injury. The exact role of NO in glomerulonephritis, however, is controversial. Results from studies attempting to inhibit excessive NO production in experimental models of glomerulonephritis are inconclusive. Whereas some studies have shown a beneficial effect of NOS inhibition on renal inflammation,8,9 detrimental effects have been reported in others.10,11
In an experimental model of anti-myeloperoxidase-associated crescentic glomerulonephritis (CGN), we recently investigated the temporal expression of NO synthases in conjunction with platelet aggregation, inflammatory cell influx, generation of reactive oxygen species (ROS), and nitrotyrosine formation. These studies showed that eNOS expression was markedly decreased and, in the early stages, was associated with extensive platelet aggregation. In addition, a marked transient induction of iNOS in polymorphonuclear cells and macrophages was observed coinciding with the generation of ROS and formation of nitrotyrosines.12 These studies suggested that NO generated by eNOS may be protective, whereas NO generated through iNOS may be involved in tissue injury in this experimental model of anti-myeloperoxidase-associated CGN. However, in a recent study, Cattell et al demonstrated that accelerated anti-glomerular basement membrane (GBM) glomerulonephritis in mice lacking iNOS is indistinguishable from that in wild-type (WT) mice, indicating that iNOS does not play an essential role in this form of glomerulonephritis in mice.13
The present study was designed to determine the role of NO radicals generated by eNOS in the development of murine accelerated anti-GBM glomerulonephritis. To this end, the course of murine anti-GBM glomerulonephritis was examined in mice homozygous for the disruption of the eNOS gene and compared to that in WT mice.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
eNOS-deficient mice were generated by homologous recombination as described previously.14 These mice were backcrossed to C57BL/6j mice for 10 generations to obtain congenic knockout mice. WT C57BL/6j mice (Jackson Laboratories, Bar Harbor, ME) were used as controls. Tail DNA from each mouse was analyzed by PCR to confirm the genotype for eNOS. All mice were bred and maintained in a specific-pathogen-free facility and received standard chow and water ad libitum.
Induction of Accelerated Anti-GBM Glomerulonephritis
Mice were immunized by i.p. injection of 0.2 mg of sheep IgG in complete Freund’s adjuvant (1:1 v/v, 200 µl total volume). After 6.5 days, mice were intravenously injected with 1 mg of sheep anti-mouse GBM IgG in phosphate buffered saline (PBS; 200 µl total volume) prepared as described previously.15 The endotoxin concentration in the stock anti-GBM preparation was 45 pg/ml. No endotoxin (<0.5 pg/ml) could be detected in the diluted anti-GBM preparation used for injection.
Albuminuria
One day after the sheep anti-GBM IgG injection and before the day of sacrifice, mice were placed in individual metabolic cages for the collection of urine. Urinary albumin content was measured with the mouse albumin enzyme-linked immunosorbent assay (ELISA) quantitation kit (Bethyl Laboratories, Montgomery, TX) using purified mouse albumin as a reference. Briefly, ELISA strips (Costar, Cambridge, MA) were coated with affinity-purified goat anti-mouse albumin in 0.1 mol/L sodium carbonate buffer, pH 9.0. Next, plates were incubated with serial dilutions of mouse urine in 1% goat serum in PBS/0.05% Tween 20 followed by peroxidase-labeled goat anti-mouse albumin. Binding was detected using tetra-methyl benzidine (TMB, Sigma, St. Louis, MO) and the reaction was stopped with 2 mol/L H2SO4. Plates were analyzed at 450 nm and albumin concentrations were determined from the standard curve and corrected for urine volume.
Anti-Sheep IgG Response
Sera collected at the time of sacrifice were tested for the presence of anti-sheep IgG antibodies by ELISA. Briefly, ELISA strips (Costar) were coated with sheep IgG (20 µg/ml) in 0.1 mol/L sodium carbonate buffer, pH 9.0. Next, strips were incubated with serial dilutions of the mouse sera starting at a 1:50 dilution in 1% donkey serum in PBS/0.05% Tween 20. Antibody binding was detected with by donkey anti-mouse IgG (Jackson Laboratories) followed by p-nitrophenyl phosphate substrate (Sigma). The optical density was analyzed at 405 nm and a standard curve was prepared from a reference serum. Antibody concentrations were computed from the linearized titration curve after log-logit transformation of the absorbances of the respective dilutions of the reference serum. The antibody concentration of the reference serum was arbitrarily set at 100 U.
Blood Urea Nitrogen Measurement
Blood urea nitrogen (BUN) was measured in sera collected at the time of sacrifice by an enzymatic degradation assay on a Vitros 250 automated analyzer (Johnson & Johnson, South Brunswick, NJ).
Light Microscopy
Renal tissue was fixed in 10% buffered formalin and embedded in paraffin using standard procedures. Two-micron sections were stained with periodic acid-Schiff (PAS) and hematoxylin and eosin (H&E). Sections were examined for glomerular hypercellularity, necrosis, thrombosis, crescents, sclerosis, interstitial leukocyte infiltrates, and fibrosis. A glomerular crescent score for each animal was determined by evaluating crescent formation in 30 glomeruli per mouse. Only glomeruli that had two or more layers of cells in Bowman’s space were considered crescentic. In addition, the presence of PAS-positive material within the glomeruli was semiquantitatively scored in 50 glomeruli per mouse according to the following scoring system: 0, absent; 1, up to one-third; 2, one-third to two-thirds; 3, more than two-thirds of the total glomerular area effaced by PAS-positive material. Results were expressed as mean scores per mouse.
Immunohistochemistry
Renal tissue was snap-frozen in isopentane and embedded in Tissue-Tek (Sakura Finetek, Torrance, CA). To determine the absence or presence of eNOS and iNOS protein, renal tissue sections were fixed in acetone and incubated with a rabbit polyclonal antibodies directed against eNOS (Santa Cruz Biotechnology, Santa Cruz, CA) or iNOS (Transduction Laboratories, Lexington, KY) After incubation with the primary antibody, sections were washed in PBS and endogenous peroxidase activity was blocked by incubation with 0.05% H2O2/PBS for 20 minutes. Next, sections were incubated with peroxidase-labeled goat anti-rabbit IgG followed by peroxidase-labeled rabbit anti-goat IgG (both from Dakopatts, Glostrup, Denmark) both diluted in 5% normal mouse serum in PBS. Binding was visualized using aminoethylcarbamizole (AEC, Sigma) and H2O2 as the substrate.
To characterize the leukocyte infiltrate, renal tissue sections (4 µm) were fixed in acetone and incubated with rat monoclonal antibodies directed against neutrophils (NIMP-R14), monocytes/macrophages (FA-11), CD11b (Pharmingen, San Diego, CA), and T cells (CD3, Pharmingen). The staining procedure was similar to the one described above using peroxidase-labeled rabbit anti-rat IgG followed by peroxidase-labeled goat anti-rabbit IgG (both from Dakopatts) both diluted in 5% normal mouse serum in PBS as secondary antibodies. Glomerular infiltrates were evaluated by counting the number of positive cells within 30 glomeruli. Numbers represent positive cells per glomerular cross section. Interstitial infiltrates were quantified on a semiquantitative scale: 0, absent; 1, minimal numbers; 2, few; 3, moderate numbers; 4, numerous.
Immunofluorescence
Four-micron renal tissue sections were fixed in acetone. The presence of sheep IgG, mouse IgG, and the deposition of fibrinogen was analyzed using fluorescein isothiocyanate-labeled rabbit anti-sheep IgG, rabbit anti-mouse IgG, and rabbit anti-human fibrinogen (cross-reactive with mouse; all from Dakopatts). The glomerular presence of sheep IgG and mouse IgG was quantitated by assessing the endpoint titer for detection of staining using four dilutions for each antibody (1:200 to 1:1600 for sheep IgG and 1:100 to 1:800 for mouse IgG). Fibrin deposits were evaluated in 30 glomeruli per mouse according to the following semiquantitative system: 0, no fibrin deposits; 1, up to one-third; 2, one-third to two-thirds; 3, more than two-thirds of the total glomerular area staining positive for fibrin. Results were expressed as mean scores per mouse.
Statistical Analysis
Results were analyzed for statistical differences by
the two-tailed Mann-Whitney U test. P 
0.05 was considered statistically significant. Values are presented as
mean ± SD.
Experimental Design
Mice between 9 and 14 weeks of age were used in this study. The following groups were studied: (i) C57BL/6J WT mice, 1 day (n = 5) and 10 days (n = 7) after induction of accelerated anti-GBM glomerulonephritis; (ii) eNOS-/- mice, 1 day (n = 5) and 10 days (n = 7) after induction of accelerated anti-GBM glomerulonephritis (from the one mouse that died at day 6 after induction of accelerated anti-GBM glomerulonephritis, only tissue for light microscopy was obtained); and normal WT (n = 4) and eNOS-/- (n = 3) mice.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both WT and eNOS-/- mice immunized with sheep IgG in complete Freund’s adjuvant developed antibodies against sheep IgG as measured by ELISA. No significant difference was found in the anti-sheep IgG titer between the two groups (arbitrary units: 252.5 ± 80.4 vs. 227.14 ± 84.9 at day 10 in eNOS-/- and WT mice, respectively). Sera from normal nonimmunized WT and eNOS-/- mice were negative.
Expression of eNOS and iNOS
Immunohistochemistry for eNOS in untreated eNOS-/- mice
confirmed the absence of eNOS protein (Figure 1A)
. In untreated WT mice, strong
immunoreactivity for eNOS was found in interstitial vessels and
glomerular and tubular capillaries (Figure 1B)
. At day 1 and 10 after
induction of anti-GBM nephritis, a decrease or absence in glomerular
eNOS expression was observed, which in most glomeruli was focal and
segmental (Figure 1, C
-E). At these time points, strong expression of
eNOS could still be observed in cortical vessels (Figure 1, C and E)
.
By immunohistochemistry, no difference in iNOS expression was detected
between WT and eNOS-/- mice. In all mice, weak immunoreactivity for
iNOS was found in tubular epithelial cells and vascular smooth muscle
cells. In addition, only occasional iNOS positive cells were detected
in interstitium and glomeruli of all mice.
|
By light microscopy, renal morphology in normal eNOS-/- mice was
indistinguishable from that in normal WT mice (Figure 2, A and B)
.
|
. All mice survived until day 10. At this
time point, a proliferative glomerulonephritis had developed with
crescent formation in 17 ± 5% of glomeruli and focal
interstitial infiltrates of leukocytes (Figure 3, A and B)
. Deposits of PAS-positive
material in the glomerular tufts, consistent with capillary thrombosis,
were focal and mild, affecting 6 to 20% (mean, 11.3 ± 5.5%) of
the glomeruli (Figure 3, A and B)
.
|
, and Figure 4
). In some glomeruli all glomerular
capillaries showed thrombus formation. Severe glomerular injury
developed by day 10, characterized particularly by deposits of
PAS-positive material in 60 to 78% (mean, 70.5 ± 8.2%) of
glomeruli (Figure 3, E and F)
. In the mice that died or became
moribund, deposits of PAS-positive material were even more extensive,
affecting more than 90% of glomeruli (mean, 94.7 ± 2.3%; Figure 3, C and D
). Also, as revealed by semiquantitative analysis, the amount
of PAS-positive material per glomerulus was significantly higher in the
eNOS-/- mice compared with WT mice (Figure 4)
. In the eNOS-/- mice,
small crescents developed in 13 ± 4% of glomeruli, which was not
significantly different from that in WT mice.
|
By immunofluorescence, strong linear glomerular deposition for
sheep IgG was found in both WT and eNOS-/- mice at days 1 and 10
after administration of sheep anti-mouse GBM IgG (Figure 5, A and B)
. The median endpoint titer
for detection was 1:1600 in both groups. Also, a prominent
granular-to-linear staining was observed for mouse IgG at both time
points in WT and eNOS-/- mice (median endpoint titer 1:400 in both
groups). At day 1 after administration of sheep anti-mouse GBM,
glomerular fibrin deposition was mild in both WT and eNOS-/- mice.
However, at 10 days, a marked increase in fibrin deposition, indicative
of extensive thrombosis, was found in eNOS-/- mice compared to WT
(fibrin score 1.93 ± 0.18 vs. 0.91 ± 0.26,
P 0.006) (Figure 5, C and D
, and Figure 6
)
|
|
In WT mice, a mild glomerular influx of PMNs and CD11b-positive
cells was found at day 1 (Figure 7)
. At
10 days, the number of glomerular CD11b-positive cells increased and a
mild influx of FA/11-positive and T cells was observed (Figure 7)
. At
this time point, the number of intraglomerular PMNs had slightly
decreased. In eNOS-/- mice, similar results were found for
CD11b-positive cells, FA/11-positive cells, and T cells. However,
glomerular influx of PMNs at day 10 was significantly higher compared
to WT mice (1.3 ± 0.12 vs. 0.77 ± 0.11,
P 0.006; normal, 0.18 ± 0.08; Figure 7
). In
the mice that became moribund, the glomerular PMN infiltrates were even
more prominent, averaging 3.2 and 2.0 PMNs per glomerular
cross-section.
|
BUN levels in normal WT and eNOS-/- mice were <20 mg/dl. In WT
mice a slight increase in BUN levels was detected at day 10 (Figure 8A)
. At this time point, eNOS-/- mice
developed severe renal impairment with markedly increased BUN levels
(105.0 ± 31.0 mg/dl vs. 47.0 ± 21.0 mg/dl,
P 0.01; Figure 8A
). In the two eNOS-/- mice
that became moribund, BUN levels were 166.0 mg/dl and 150.0 mg/dl.
|
; albuminuria in normal WT
(n = 4) and eNOS-/- mice
(n = 3) was <50 µg/18 hours). However, no
significant differences were found between eNOS-/- and WT mice at any
time point. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Over the past few years, studies have pointed to a complex role of NO in renal inflammation.6 Increased production of NO radicals has been demonstrated in several experimental models of glomerulonephritis.6 In most, if not all, cases the increased generation of NO radicals has been attributed to the induction of iNOS in infiltrating inflammatory cells or intrinsic renal cells such as mesangial cells. In these models of glomerulonephritis, iNOS mRNA can be detected in glomeruli by ribonuclease protection assays and reverse transcription PCR, and iNOS protein by immunohistochemistry. Increased production of NO radicals can contribute to renal injury through direct cytotoxic actions. In addition, NO radicals can react with superoxide radicals generating the highly reactive compound peroxynitrite.16 Peroxynitrite has been shown to initiate lipid peroxidation and induces nitration of tyrosine residues, leading to loss of protein structure and function.16
To further define the role of increased NO production in renal inflammation, several studies have attempted to decrease excessive NO production either by using inhibitors of NO synthases or by depletion of circulating arginine. In some of these studies, inhibition of NO production has been shown to augment glomerulonephritis, implicating a toxic role for NO radicals.8,9,17 In other studies, however, blockade of NO production aggravated renal injury, pointing to a protective role of NO radicals.10,11 Until recently, the major NOS inhibitors lacked NOS isoform specificity, which may explain part of the conflicting data. Overall, these studies have indicated that increased NO production during renal inflammation may have detrimental or protective effects, which probably depends on the NOS isoform generating NO radicals, the amount of NO radicals produced, and the simultaneous production of ROS. The present study clearly demonstrates that inactivation of eNOS by disruption of the eNOS gene exacerbates renal injury in the accelerated anti-GBM glomerulonephritis model. As such, these results provide direct evidence for a protective role of eNOS derived NO radicals in renal inflammation.
The exact mechanisms by which lack of eNOS-derived NO radicals exacerbates renal injury in this model are not clear, but several possibilities may be considered.
Mice lacking a functional gene for eNOS were originally generated to study the involvement of NO radicals produced by eNOS in blood pressure regulation.14,18 These studies showed that eNOS-/- mice were indistinguishable from WT mice in general appearance and histology, but had significantly increased blood pressures associated with a twofold increase in plasma renin activities.14 Also, long term treatment with the nonselective NOS inhibitors L-NAME or L-NNA in rats causes hypertension and leads to proteinuria and renal morphological changes, such as capillary collapse and tubulointerstitial injury.19,20 In addition, hypertension is a factor known to exacerbate glomerulonephritis in rats.21 Thus, the increased blood pressures in eNOS-/- mice may have contributed to the increased renal injury after induction of accelerated anti-GBM glomerulonephritis we observed; this possibility warrants further investigation.
In vitro studies have also demonstrated that endothelial cell-derived NO radicals inhibit platelet aggregation and platelet and leukocyte adhesion to the endothelium.3 In vivo, administration of the nonselective NOS inhibitor L-NAME in models of endotoxemia exacerbates liver and kidney injury associated with increased thrombosis and leukocyte infiltration, especially PMNs.11,22,23 Part of these effects have been attributed to increased adhesion molecule expression (ICAM-1, P-selectin) by the endothelium after inhibition of endogenous NO production most likely derived from eNOS.22-24 In the present study, glomerular capillary thrombosis was a prominent feature of the disease in eNOS-/- mice. In addition, a small but significant increase in glomerular PMN influx was observed in the eNOS-/- mice at day 10.
In conclusion, the course of accelerated anti-GBM glomerulonephritis is severely aggravated in mice lacking NO production by eNOS. These results point to a protective role of eNOS-derived NO production in glomerulonephritis and emphasize its importance in maintaining the antithrombogenic and anti-inflammatory properties of the endothelium.
|
Footnotes |
|---|
Supported by a grant from the Dutch Kidney Foundation (C97–1682).
Accepted for publication October 27, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Nakayama, W. Sato, T. Kosugi, L. Zhang, M. Campbell-Thompson, A. Yoshimura, B. P. Croker, R. J. Johnson, and T. Nakagawa Endothelial injury due to eNOS deficiency accelerates the progression of chronic renal disease in the mouse Am J Physiol Renal Physiol, February 1, 2009; 296(2): F317 - F327. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Glushakova, T. Kosugi, C. Roncal, W. Mu, M. Heinig, P. Cirillo, L. G. Sanchez-Lozada, R. J. Johnson, and T. Nakagawa Fructose Induces the Inflammatory Molecule ICAM-1 in Endothelial Cells J. Am. Soc. Nephrol., September 1, 2008; 19(9): 1712 - 1720. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. J. Lowenstein Nitric oxide regulation of protein trafficking in the cardiovascular system Cardiovasc Res, July 15, 2007; 75(2): 240 - 246. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. A. Asgeirsdottir, J. A. A. M. Kamps, H. I. Bakker, P. J. Zwiers, P. Heeringa, K. van der Weide, H. van Goor, A. H. Petersen, H. Morselt, H. E. Moorlag, et al. Site-Specific Inhibition of Glomerulonephritis Progression by Targeted Delivery of Dexamethasone to Glomerular Endothelium Mol. Pharmacol., July 1, 2007; 72(1): 121 - 131. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Nakagawa Uncoupling of the VEGF-endothelial nitric oxide axis in diabetic nephropathy: an explanation for the paradoxical effects of VEGF in renal disease Am J Physiol Renal Physiol, June 1, 2007; 292(6): F1665 - F1672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Hamik, Z. Lin, A. Kumar, M. Balcells, S. Sinha, J. Katz, M. W. Feinberg, R. E. Gerszten, E. R. Edelman, and M. K. Jain Kruppel-like Factor 4 Regulates Endothelial Inflammation J. Biol. Chem., May 4, 2007; 282(18): 13769 - 13779. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Forbes, B. A. Thornhill, M. H. Park, and R. L. Chevalier Lack of Endothelial Nitric-Oxide Synthase Leads to Progressive Focal Renal Injury Am. J. Pathol., January 1, 2007; 170(1): 87 - 99. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Lang, D. Benke, F. Eitner, D. Engel, S. Ehrlich, M. Breloer, E. Hamilton-Williams, S. Specht, A. Hoerauf, J. Floege, et al. Heat Shock Protein 60 Is Released in Immune-Mediated Glomerulonephritis and Aggravates Disease: In Vivo Evidence for an Immunologic Danger Signal J. Am. Soc. Nephrol., February 1, 2005; 16(2): 383 - 391. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chu and H. G. Bohlen High concentration of glucose inhibits glomerular endothelial eNOS through a PKC mechanism Am J Physiol Renal Physiol, September 1, 2004; 287(3): F384 - F392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. SenBanerjee, Z. Lin, G. B. Atkins, D. M. Greif, R. M. Rao, A. Kumar, M. W. Feinberg, Z. Chen, D. I. Simon, F. W. Luscinskas, et al. KLF2 Is a Novel Transcriptional Regulator of Endothelial Proinflammatory Activation J. Exp. Med., May 17, 2004; 199(10): 1305 - 1315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Kruger, D. Benke, F. Eitner, A. Lang, M. Wirtz, E. E. Hamilton-Williams, D. Engel, B. Giese, G. Muller-Newen, J. Floege, et al. Identification and Functional Characterization of Dendritic Cells in the Healthy Murine Kidney and in Experimental Glomerulonephritis J. Am. Soc. Nephrol., March 1, 2004; 15(3): 613 - 621. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Li, C. W. Yang, J. H. Park, S. W. Lim, B. K. Sun, J. Y. Jung, S. B. Kim, Y. S. Kim, J. Kim, and B. K. Bang Pravastatin treatment attenuates interstitial inflammation and fibrosis in a rat model of chronic cyclosporine-induced nephropathy Am J Physiol Renal Physiol, January 1, 2004; 286(1): F46 - F57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Attia, R. Goldschmeding, M. A. Attia, P. Boer, H. A. Koomans, and J. A. Joles Male gender increases sensitivity to renal injury in response to cholesterol loading Am J Physiol Renal Physiol, April 1, 2003; 284(4): F718 - F726. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Suganami, M. Mukoyama, A. Sugawara, K. Mori, T. Nagae, M. Kasahara, K. Yahata, H. Makino, Y. Fujinaga, Y. Ogawa, et al. Overexpression of Brain Natriuretic Peptide in Mice Ameliorates Immune-Mediated Renal Injury J. Am. Soc. Nephrol., December 1, 2001; 12(12): 2652 - 2663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. SHAO, T. MIYATA, K. YAMADA, N. HANAFUSA, T. WADA, K. L. GORDON, R. INAGI, K. KUROKAWA, T. FUJITA, R. J. JOHNSON, et al. Protective Role of Nitric Oxide in a Model of Thrombotic Microangiopathy in Rats J. Am. Soc. Nephrol., October 1, 2001; 12(10): 2088 - 2097. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
