| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Departments of Discovery Technologies (N.B., C.C., S.Z.) and Transplantation Research (R.H., J.L., C.P., C.B.), Novartis Institutes for BioMedical Research, Lichtstrasse 35, WSJ-386.2.09, CH-4002 Basel, Switzerland. Received May 25, 2005; revision requested July 20; revision received September 2; accepted September 22; final version accepted November 23. Address correspondence to N.B. (e-mail: nicolau.beckmann{at}novartis.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: Experiments followed the Swiss federal regulations of animal protection. Male Fisher (n = 37) and Lewis (n = 77) rats were used. After removal of a native recipient kidney and transplantation of a donor kidney, the recipient rat's contralateral kidney was removed. Allografts and control syngeneic grafts comprised, respectively, kidneys from Fisher and Lewis donors transplanted into Lewis rats. Recipients were imaged by using a gradient-echo MR sequence 24 hours after intravenous administration of superparamagnetic iron oxide (SPIO) particles. Biochemical analyses of blood and urine, as well as assessments of Banff scores (reference standard for histologic classification of graft rejection), were performed. Statistical tests used were analysis of variance for multiple comparisons with Bonferroni tests, Mann-Whitney tests, and Pearson correlations with Bonferroni corrections.
Results: A SPIO dose–dependent decrease in cortical MR signal intensity occurred in allografts between 8 and 16 weeks after transplantation. A strong significant negative correlation (P = .005 for 0.3 mL/kg SPIO dose, P = .003 for 1.0 mL/kg SPIO dose) was found between MR signal intensity and Banff scores, which deteriorated over the experimental period. Proteinuria occurred at 16 weeks. Blood and urine creatinine levels remained unchanged up to week 28.
Conclusion: This MR imaging method is more robust than the usually adopted creatinine clearance method for the detection of early signs of allograft chronic rejection in the Fisher-to-Lewis rat kidney transplantation model.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
The development of therapeutic strategies for the care of patients with transplants depends critically on disease-relevant animal models (5). In the case of kidney transplantation, hallmarks of chronic allograft rejection that mimic aspects seen in human graft rejection (vascular and tubular remodeling, mononuclear cell infiltration, proteinuria) can be induced in rats in the life-supporting Fisher-to-Lewis kidney transplantation model (Fisher 344 rats as donors; Lewis rats as recipients) when the recipients are treated for 10 days after surgery with a low dose of cyclosporine (6–8).
In view of early translational research toward clinical application, it is of interest to have access to noninvasive monitoring at the preclinical level, allowing repetitive assessment of parameters in the same animal. Magnetic resonance (MR) imaging has proved to be an attractive approach to following the development of chronic allograft rejection in transplantation models in rodents (9). Anatomic and perfusion MR imaging (10–13), as well as MR renography (14), have been successfully applied to follow the status of kidney grafts in rats. Detection of macrophage infiltration into grafts by labeling the cells with iron oxide–containing nanoparticles has also been performed to characterize acute and chronic kidney allograft rejection in rats (14,15). Superparamagnetic iron oxide (SPIO) particles with a diameter of up to 150 nm can be taken up by macrophages through endocytosis (16), leading to a decrease in MR imaging signal intensity in organ regions where the macrophages accumulate.
The purpose of our study was to evaluate detection of iron-loaded macrophages at MR imaging as a noninvasive method of monitoring early signs of chronic allograft rejection in the life-supporting Fisher-to-Lewis rat kidney transplantation model.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Animals
Thirty-seven male Fisher 344 rats (genetic strain, RT1lv1) (Iffa-Credo, L'Arbresle, France) and 77 male Lewis rats (genetic strain, RT1l) (Harlan-Olac, Zeist, the Netherlands) weighing approximately 250 g were used. They were housed in a temperature- and humidity-controlled environment and had free access to standard rat chow and tap water. An acclimatization period of at least 1 week was allowed before transplantation.
Experimental Groups
Life-supporting allogeneic kidney grafts.—Thirty-seven Lewis rat recipients of kidneys from Fisher rats were examined. Animals were treated with 1.5 mg of cyclosporine (Neoral; Novartis Pharma, Basel, Switzerland) per kilogram of body weight per day by mouth (by means of gavage) for 10 days after transplantation.
Fifteen recipients were examined with MR imaging 8, 12, and 16 weeks after transplantation. Biochemical analyses of blood and urine were performed at the same time points. Histologic analysis was performed at week 16 after MR imaging and biochemical analyses. These animals received SPIO particles, as described in the SPIO Contrast Medium section below.
Biochemical analyses of blood and urine were performed in another 22 recipients 8, 12, 16, 22, and 28 weeks after transplantation. Histologic analysis was performed at week 8 (in five rats), week 16 (in three rats), and week 28 (in five rats).
Life-supporting syngeneic kidney grafts.—Fifteen Lewis rat recipients of kidneys from Lewis rats (control group) were examined with MR imaging at weeks 8, 12, and 16 after transplantation. Animals were treated with 1.5 (mg · kg–1)/d cyclosporine by mouth (by means of gavage) for 10 days after transplantation. These animals received SPIO particles, as described in the SPIO Contrast Medium section below.
Kidneys of normal Lewis rats.—Ten Lewis rats underwent MR imaging three times, at intervals of 4 weeks, and served as additional control rats. These animals received SPIO particles, as described in the SPIO Contrast Medium section below.
Surgical Procedure
For pain treatment, animals received subcutaneous 0.03 mg/kg buprenorphine (Temgesic; Essex Chemicals, Lucerne, Switzerland) just before transplantation and 6 hours after recovery. Kidney transplantation was performed by one author (J.L., who had 9 years of experience with rat transplantation procedures) with rats under general anesthesia, which was induced with 2-vol% (2% of volume of gas administered) isoflurane (Forene; Abbott, Baar, Switzerland) in a 1:1 mixture of O2 and room air administered through a nose cone. Before being removed from the donor, the kidney was perfused for 3 minutes with University of Wisconsin solution. Kidneys were preserved for 24 hours in cold (4°C) University of Wisconsin solution before transplantation. The grafts were not perfused ex vivo. After the recipient's kidney was removed, the donor kidney was placed in position with end-to-end anastomoses of the donor's renal artery to the recipient's renal artery by using 10-0 Ethilon (Delasco, Council Bluffs, Iowa) interrupted sutures and of the donor's renal vein to the recipient's renal vein by using 10-0 Ethilon continuous sutures. The donor and recipient ureters were anastomosed end-to-end with 10-0 Ethilon interrupted sutures. The transplanted kidneys had life-supporting function—that is, nephrectomy of the contralateral kidney was performed at transplantation time.
MR Imaging
MR imaging measurements were performed by two authors (N.B. and S.Z.) with a small-animal MR imaging unit (Biospec 47/40; Bruker Medical Systems, Ettlingen, Germany) operating at 4.7 T. A 7-cm-diameter birdcage resonator was used for excitation and detection. During MR imaging measurements, rats were anesthetized with 2-vol% Forene in a 1:2 mixture of O2 and N2O that was administered through a nose cone. Body temperature was maintained at 36.5°C ± 1 by a flow of warm air that was regulated by a rectal temperature probe (DM 852; Ellab, Copenhagen, Denmark). Animals respired spontaneously during image acquisition, and no respiratory gating was applied.
Anatomic images were acquired by using a gradient-echo sequence (17) with the following parameters: repetition time msec/echo time msec, 16.8/8.4; bandwidth, 30 kHz; flip angle, approximately 10°; field of view, 6 x 6 cm2; matrix size, 256 x 128; section thickness, 1.5 mm; and number of signals acquired, 20. Use of these parameters resulted in a measurement time of 43 seconds per image. A single section image was obtained by interpolating the data set to 256 x 256 pixels. For every animal and at each time point, five to seven sections were acquired sequentially with a displacement corresponding to a section thickness; a substantial portion of the graft was thus imaged.
For each section, the average signal intensity was computed for a region of interest that was drawn manually to encompass the kidney cortex. The signal intensity in abdominal muscle served as a reference. Thus, for each section, we reported the signal intensity in the kidney cortex as relative to that in the muscle to exclude the possibility that signal drift could influence the results. For each animal and time point, the mean renal cortex signal intensity (corrected for the muscle signal intensity) was determined from these five to seven sections. The author who performed these analyses (S.Z.) was blinded to the origin of the data sets.
SPIO Contrast Medium
Colloid-based SPIO (Endorem [11.2 mg of iron per milliliter]; Guerbet, Aulnay-sur-Bois, France) with a mean particle size of 150 nm and R2 and R1 relaxivities of 160 and 40 mmol–1 · sec–1 (18), respectively, were used. Animals received the SPIO particles (0.3 or 1.0 mL/kg) intravenously as a bolus (administration time, 2–3 seconds) 24 hours before each imaging session.
Urine Biochemical Values
Urine secretion was assessed by two investigators (C.P. and R.H.) over a period of 24 hours by placing the rats in individual metabolic cages in which the bottom was permeable. Samples were collected in Plexiglas tubes, and the creatinine and total protein content in the urine were assessed by using a Synchron CX5 instrument (Beckman Coulter, Fullerton, Calif). The glomerular filtration rate was determined by multiplying the ratio of creatinine in urine and in plasma by the amount of urine excreted per hour.
Blood Biochemical Values
Blood was taken from the sublingual vein into edetic acid–coated tubes for monitoring hematologic values by using a counter (Technicon H*1E; Bayer, Zurich, Switzerland). Blood was also collected in heparin-coated Eppendorf tubes and centrifuged for 3 minutes at 13 000 rpm in an Eppendorf tabletop centrifuge. The resulting supernatant was biochemically analyzed (with Synchron CX5 equipment). Two authors (C.P. and R.H.) assessed plasma creatinine and urea values.
Histologic Examination
Animals were sacrificed by means of inhalation of carbon dioxide immediately after an MR imaging examination. The kidneys, liver, and spleen were removed by one author (C.C.) and were fixed with 10% phosphate-buffered formalin (pH 7.2) for approximately 72 hours. After fixation, kidneys were cut longitudinally to include the cortex and medulla, as well as the renal papilla. A transverse slice through the left lobe of the liver and a transverse slice through the spleen were also processed. The tissues were dehydrated with graded concentrations of ethanol (50%, 70%, 80%, 95%, 100%) and were embedded in paraffin. The blocks were cut serially at a thickness of 3 µm.
Histologic analysis and assessment of Banff scores.—The slides were stained by using (a) hematoxylin-eosin to enable assessment of the degree of cellular infiltration and tubular atrophy, (b) the periodic acid–Schiff reaction to enable detection of the extent of glomerulosclerosis and tubular infiltration (a minimum of 100 glomeruli were counted in each section, and the number of sclerosed glomeruli was expressed as a percentage of the total number of glomeruli), (c) the Verhoeff reaction to enable assessment of the degree of vasculopathy, and (d) acid–fuchsin-orange G staining to enable assessment of the extent of interstitial fibrosis. Vasculopathy was defined as endothelial thickening and/or disruption of the internal lamina elastica. The percentage of diseased to total vessels was calculated for each kidney slice.
Renal structure damage was scored semiquantitatively for interstitial cellular infiltration, tubulopathy, glomerulopathy, and vasculopathy by using the Banff criteria (19–21). Each of the four measures of renal structural damage was scored according to the following scheme: a score of 0 indicated 0%–10% change; a score of 1, 11%–25% change; a score of 2, 26%–50% change; and a score of 3, greater than 50% change. A summed Banff score was calculated for each slice by adding the Banff scores of the four measures of renal damage. The degree of allograft rejection was considered to be mild, moderate, and severe for summed Banff scores of 1–4, 5–8, and 9–12, respectively. The examiner (C.C.), a pathologist with 25 years of experience in histologic analysis, was blinded to the origin of the slices.
Assessment of ED1 labeling in kidney grafts.—Slides were stained with the monoclonal antibody ED1 (Serotec, Düsseldorf, Germany), which is directed against rat macrophage lysosomal membrane. Morphometric analyses were performed by using software (Histolab; Microvision Instruments, Evry, France). Stained slides were examined with a light microscope (Eclipse E600; Nikon Micro Science, Egg, Switzerland) connected to a three color code digital video camera (DXC-970MD; Sony Austria, Vienna, Austria). After capture of a representative frame (approximately 5 700 000 µm2) that contained approximately 16 glomeruli, the color corresponding to ED1 staining was extracted by threshold setting, and the area was automatically calculated. Results were expressed as a percentage of ED1-labeled cells to total measured surface. Analyses were performed by one investigator (C.C.), who has 15 years of experience with immunohistochemistry.
Detection of iron in kidneys.—Iron was detected in the cortical area of the tissue after the slides were stained by using the Perls Prussian blue reaction for iron. Stained slides were examined with the Eclipse E600 light microscope, which was equipped with a motorized stage and connected to the DXC-970MD three color code digital video camera. The iron content was determined by one author (C.C.) using Histolab. The color corresponding to iron was extracted by means of threshold setting, which allowed the amount of iron in the cortical area to be automatically calculated. Results were expressed as percentage of iron to the total cortical surface of the kidney.
Detection of iron in liver and spleen.—The same equipment and the same staining methods described above were used by the same investigator to detect iron in the liver and spleen. After a representative frame was captured, the color corresponding to iron staining was extracted by means of threshold setting, and the area was automatically calculated. The results were expressed as percentage of iron to the total organ surface.
Statistical Analysis
Analysis of variance for multiple comparisons with Bonferroni tests were performed for the MR imaging signal intensities and the biochemical values. Mann-Whitney tests were used to compare the Banff scores at different time points after transplantation. Pearson correlations with Bonferroni correction were used for comparing the Banff scores with the MR imaging signal intensities and the amount of protein in the urine. A P value of less than .05 was considered to indicate a statistically significant difference. One author (N.B.) used statistical software to perform analysis of variance tests and Pearson correlations (SigmaStat, version 3.1, 2003; Systat Software, Point Richmond, Calif) and Mann-Whitney tests (SYSTAT, version 10.2, 2004; Systat Software).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
For the Fisher-to-Lewis model, it had been shown previously that decreases in signal intensity in the kidney cortex reflecting the influx of iron-loaded macrophages in recipients that had received SPIO manifested substantially earlier than did the impairment of kidney function assessed at MR imaging (14). Despite those promising results, a thorough comparison with Banff scores and blood and urine biochemical parameters, as performed in our study, was essential to qualify the approach. The relative cortical MR imaging signal intensity in the allografts decreased progressively between 8 and 16 weeks after transplantation, the decrease being dependent on the SPIO dose. The progressive deterioration of the kidney structure that was revealed at histologic examination suggests that the course of the cortical signal intensity reflected the progressive rejection process of the grafts. Indeed, the relative cortical signal intensity in the grafts was significantly negatively correlated to the Banff scores at week 16. From the biochemical values determined in samples taken from the same rats that had been imaged with MR, only protein in the urine was significantly elevated, reflecting proteinuria, which is a characteristic of the allograft rejection model adopted here (22,23). The other biochemical values were unchanged between 8 and 28 weeks after transplantation. These observations indicate that the MR imaging signal intensity in the cortex after administration of SPIO is a more sensitive marker for the early noninvasive characterization of chronic rejection in the Fisher-to-Lewis model of kidney allotransplantation than the commonly used creatinine clearance method. The correlation between cortical MR signal intensity and results of histologic evaluation was more consistent and stronger than that between signal intensity and protein excretion.
The presence of macrophages mainly in the cortex is a feature of inflammatory processes associated with graft rejection. After kidney transplantation, acute inflammatory changes may occur in the glomeruli (glomerulitis, defined as an increase in the number of mononuclear cells in the glomerular capillary lumina, often accompanied by reactive changes and swelling of the endothelial cells). Moderate or severe glomerulitis occurs in about 15% of graft biopsies from the early posttransplantation period in humans (24). In the Fisher-to-Lewis model, a few monocytes and/or macrophages bind to glomeruli and vessels at 8 weeks after transplantation; by 12 weeks, binding to glomeruli is high (25). However, increasing numbers of monocytes and/or macrophages in kidney allografts peak at 16 weeks, and are localized preferentially in glomeruli, where interleukin 1, interleukin 6, and tumor necrosis factor–
expression also becomes intense and correlates with progressive glomerulosclerosis (25). Results of several studies (26–28) support the hypothesis that macrophage-derived inflammation is a cofactor for chronic allograft rejection, and monocytes and/or macrophages and T cells are the predominant graft-invading cells of rat renal allografts with chronic rejection (29–31).
Macrophage trafficking through the allograft is a dynamic process that is present at all stages in the life of a transplanted solid organ. An initial influx of macrophages follows the ischemia-reperfusion injury of transplant surgery. In allografts that are not being rejected, macrophages may be present in low numbers. Macrophages are present in large numbers during episodes of allograft rejection and are also prominent during the slow death of the organ through chronic rejection. Accumulation of macrophages within the allograft reflects a balance of blood monocyte recruitment and subsequent local proliferation versus macrophage apoptosis or exit from the organ (32–34). Following the reasoning of our work, one would expect that, for a given dose of contrast agent, the more dramatic the rejection, the lower the signal intensity, because more macrophages would infiltrate the graft. Thus, in the case of acute rejection, or pronounced chronic rejection, the decreases in signal intensity should be more dramatic than those reported here. Although this reasoning is plausible, comparisons between acute and chronic rejection still need to be carefully addressed. Especially in the case of acute rejection, the picture may be complicated by several factors, including surgical trauma, vascular changes (rapid neointima formation may hamper the delivery of contrast agent to the graft), hemorrhages, and a high macrophage turnover.
A potential limitation of the method is that if SPIO particles are administered too often, saturation of the MR imaging signal could occur. In cases where the contrast agent is applied (for example) every few days, it would be important to determine at histologic examination if there is free contrast agent. It cannot be excluded that some SPIO particles could leak out from the vessels and not be phagocytosed by macrophages. Such particles could then lead to excessive decreases in signal intensity, especially in view of impaired kidney function that might potentially cause excretion problems. Also, because we used a gradient-echo sequence without respiratory gating, we observed artifacts caused by susceptibility differences in approximately 5% of the images. Finally, we limited the biochemical analyses to assessments of creatinine (in urine and blood), total protein (in urine), and urea (in blood). It might be the case that more sensitive markers of chronic rejection could be detected by using refined proteomics tools.
In conclusion, our results demonstrate that detection of iron-labeled macrophage infiltration with MR imaging is a more sensitive approach for investigating early signs of allograft chronic rejection in the Fisher-to-Lewis kidney transplantation model than the routinely adopted creatinine clearance assay. Detection in situ of early changes associated with allograft rejection can have an effect in preclinical studies by facilitating the investigation of immunomodulatory therapies for transplantation (35–37) and by shortening the duration of the experimental period. Furthermore, it cannot be excluded that the approach may be translated into the clinic in the near future.
Practical application: We routinely use the MR imaging approach described here to test compounds in the Fisher-to-Lewis model of kidney transplantation. Because for this model, the MR imaging readout has been extensively validated and compared with the Banff scores, the effect of the technique on the reduction of the experimental duration of preclinical studies is twofold: (a) No further histologic analysis is necessary for evaluating drug effects, and (b) because MR imaging is more sensitive than the previously used creatinine assessments, the studies can be substantially shortened.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: SPIO = superparamagnetic iron oxide
See Materials and Methods for pertinent disclosures.
Author contributions: Guarantor of integrity of entire study, N.B.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, N.B.; experimental studies, all authors; statistical analysis, N.B., R.H.; and manuscript editing, N.B., J.L., C.B.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Lewis rats: hemodynamics, macrophages, and cytokines. Kidney Int 2000;57:2618–2625.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
