 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pathophysiology of Renal Disease and Progression |
Departments of * Histopathology and Renal Medicine and Imaging Sciences Department, Ultrasound Group, MRC Clinical Sciences Centre, Imperial College, and Histopathology Unit, London Research Institute, Cancer Research UK, London, United Kingdom
Address correspondence to: Dr. Candice Roufosse, Department of Histopathology, Hammersmith Campus, Imperial College, DuCane Road, W12 0NN, London, United Kingdom. Phone: +44-20-7269-3434; Fax: +44-20-7269-3491; E-mail: c.roufosse{at}imperial.ac.uk
Received for publication August 1, 2005. Accepted for publication December 29, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
-galactosidase reporter molecules under the control of a 17-kb promoter and enhancer element of the gene encoding the 
2 chain of the collagen I was used. Male transgenic BM was transplanted into female wild-type C57BL/6 mice (n = 14), and unilateral ureteric obstruction was performed later to induce renal fibrosis. In the obstructed kidney of the BM-chimeric female mice, a mean of 8.6% of smooth muscle actin–positive interstitial cells were Y chromosome positive. Increased collagen I mRNA in the obstructed kidney was detected by in situ hybridization. No luciferase activity was detected by enzyme assays in tissue homogenates of BM recipients, and very few luciferase mRNA transcripts were seen, mainly in tubular cells. -Galactosidase activity was not a useful reporter molecule because it could not be distinguished from enhanced endogenous -galactosidase activity in the obstructed kidney. These results indicate that BM-derived interstitial cells do not make a significant contribution to collagen I synthesis in the context of renal injury. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
The renal interstitium is a normally inconspicuous space that contains extracellular matrix (ECM) within which are closely apposed fibroblasts, vascular pericytes, and inflammatory cells (3). Under normal circumstances, very little ECM is produced by these cells (4). After injury, fibroblasts undergo a process called activation, which involves proliferation and excessive production of ECM components, including collagens I and III (5). Some activated fibroblasts, termed myofibroblasts, also acquire smooth muscle actin (SMA) expression, which is thought to confer a degree of mobility and the ability to contract scar tissue. Myofibroblasts and fibroblasts both have been incriminated in the excess ECM production in kidney fibrosis (6,7).
The origin of the increased numbers of myofibroblasts and fibroblasts in fibrosis is still uncertain. A number of different possibilities have been considered. Cohnheim’s original studies on wounding in 1867 (cited in [8,9]) showed that inflammatory cells entered a wound from the circulating blood, and he further suggested that these inflammatory cells then may give rise to the wound fibroblasts. This hypothesis was tested during the following century, in particular elegantly by Ross et al. (10). Using a parabiotic rat model, they found that skin wound fibroblasts were not derived from a population of tritiated thymidine–labeled bone marrow (BM) cells. By exclusion, Ross et al. hypothesized that wound fibroblasts were derived from the proliferation of local cells. In the kidney, another possible origin for fibroblasts is epithelial-mesenchymal transition (EMT), whereby tubular cells convert to interstitial fibroblasts (11–13). This theory is based on in vitro and in vivo observations that tubular epithelial cells after certain stimulations are able to produce ECM, lose epithelial markers, and acquire markers of fibroblastic differentiation (14–17).
Finally, recent investigations in the kidney have suggested a possible origin of fibroblasts from the BM via the peripheral blood (11,18–21). BM-derived cells that were identified by their chromosomal content (male, Y chromosome) or the expression of a reporter molecule such as enhanced green fluorescence protein were found in the interstitium of fibrosed kidneys, amounting to up to 30% of the population of fibroblasts. Using similar BM cell-tracking methods, other investigators have found BM-derived (myo)fibroblasts in a number of organs, including the lung (22), skin (23), liver (24), and tumor stroma (25).
These various origins of fibroblasts and myofibroblasts in renal interstitial fibrosis are not mutually exclusive. Determining the contribution of each modality to fibrosis, however, may suggest possible routes for therapeutic intervention. In view of the current development of stem cell therapies that involve injection of adult BM cells into organs to enhance repair (26,27), we undertook to investigate further the potential contribution of BM-derived cells to renal scarring. In particular, we sought to determine whether BM-derived cells could actively participate in ECM production. We used a transgenic mouse line that expresses two reporter molecules (luciferase and -galactosidase) under the control of the promotor and enhancer elements of the collagen I (2 chain) gene (28). We transplanted BM from these transgenic mice into wild-type mice, induced kidney fibrosis, and looked for reporter gene expression in the injured kidney as a surrogate marker of collagen synthesis by BM-derived cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
2 chain of pro-collagen type I) promoter, which consists of 17 kb 5' of the transcription start site, including the far upstream enhancer region, fused to luciferase and -galactosidase reporter genes as described previously (28). This was bred onto a C57BL/6 background over six generations. In this transgenic mouse, transgene expression mirrors that of the endogenous pro-COL1A2 gene in development of and after injury of adult tissues (29,30). Age-, strain-, and sex-matched wild-type mice were used in this experiment. C57BL/6 mice were obtained from Harlan UK animal suppliers (Oxon, UK). Mice were kept in a clean environment, and experiments were performed according to institutional and UK Home Office guidelines.
Sex-Mismatched BM Transplantation
BM from the femurs, tibiae, and humeri of transgenic male mice was flushed and resuspended in RPMI, 10% FCS, 100 U/ml penicillin, and 100 µg/ml streptomycin. Female C57BL/6 mice (n = 14) were irradiated at 8 Gy using gamma rays from a cesium source irradiator. Within 2 h of irradiation, 10 million donor cells were injected via a tail vein. Mice that received a transplant were housed in individually ventilated cages. As control animals, wild-type female BM was transplanted into wild-type female mice (n = 4), and male transgenic BM was transplanted into transgenic male mice (n = 4).
Reconstitution of hematopoietic cells was checked in a subset of mice that received a transplant (n = 4) using in situ hybridization (ISH) for the mouse Y chromosome on BM cytospins. Female and male mice were used as negative and positive controls. Femoral bones were flushed with PBS, and approximately 10,000 cells were spun onto slides for 5 min at 800 x g. The slides then were fixed for 10 min in 4% paraformaldehyde. After a brief digestion in pepsin 0.4%/0.1 M HCl at 37°C for 30 s to 2 min, ISH was performed as for tissue sections described below. Irradiation dose and quantity of cells injected were also chosen according to previous publications showing a high proportion of donor mesenchymal stem cells in the BM 1 mo after transplantation (31).
BM Cell Culture
Mononuclear cells from whole BM of transgenic mice were separated using a Ficoll gradient and were plated at 5000 cells/cm2 in a MesenCult Basal Medium supplemented with Mesenchymal Stem Cell Stimulatory Supplements (StemCell Technologies Inc., Meylan, France). Nonadherent cells were removed at 72 h. Cells were trypsinized when confluent and plated at 5000 cells/cm2 in the same medium (32). Cells were grown in six-well plates for imaging using IVIS Imaging System 100 Series (Xenogen Corp., Alameda, CA) and on BD Falcon glass culture slides (BDH Laboratory Supplies, Inc., Hertfordshire, UK) for in situ hybridization. Culture slides were fixed for 15 min in neutral-buffered formalin before in situ hybridization.
Unilateral Ureteric Obstruction
Fibrosis was induced in one kidney by unilateral ureteric obstruction (UUO) 6 wk after BM transplantation. BM-chimeric and control animals underwent the same procedure.
Animals were anesthetized with a mixture of isoflurane and oxygen. Two ties were knotted around the mid portion of the left ureter, using a thin nonabsorbable suture (5/0, Mersilk). Animals were killed at 7 d or 14 d after UUO.
The amount of cortical fibrillary collagen was assessed by observing sections stained with picrosirius red (33) under double-polarized light using an Olympus BX51 microscope. Three fields (x100 magnification) of renal cortex were captured with a 3-CCD JVC Digital Camera (KY F75U). Images were converted to gray-scale 256-bit images, then analyzed using Image Pro-Plus 5.0 Software (Media Cybernetics UK, Wekingham, UK). Large vessels were manually excluded from analysis. The percentage of cortical area that contained picrosirius red–stained material was measured for each field (34), and the mean values for three fields were calculated. The values for control and obstructed kidneys were compared between wild-type and BM-chimeric mice.
Reporter Molecule Assays
Samples from obstructed and contralateral nonobstructed kidneys were snap-frozen in liquid nitrogen, then homogenized on ice. The cell lysate was used to measure luciferase and -galactosidase with a chemiluminescent reporter gene assay system (Dual-Light, Applied Biosystems). Protein in the sample was measured using BCA protein kit (Pierce, Rockford, IL). The results were expressed per milligrams of protein.
Luciferase production by cultured transgenic BM cells was assessed using IVIS Imaging System 100 Series, which measures bioluminescence that is produced by luciferase enzyme activity and superimposes the bioluminescence image on a standard black and white reference image of the system analyzed. Luciferin was added to cell culture wells at a concentration of 225 µg/ml. Within 1 min, the cell culture plate was read using an exposure time of 5 min. Wild-type C57BL/6 mesenchymal BM cells were used as a negative control (courtesy of Tzung-Chih Tang, Department of Hematology, Hammersmith Campus, Imperial College, London, UK).
Histology
Tissues from all killed animals were fixed in neutral-buffered formalin for 24 h then transferred to 70% ethanol. Fixation time was controlled carefully to standardize digestion times for in situ hybridization. Tissues were analyzed for the presence of tubular damage, glomerular damage, and tubulointerstitial fibrosis using hematoxylin and eosin, periodic acid-Schiff, and picrosirius red stains.
ISH for Collagen I and Luciferase
ISH for collagen 2(I) ( 2 chain of collagen I) and luciferase mRNA was carried out on blocks of formalin-fixed paraffin-embedded tissues and on mesenchymal stem cells cultured on BD Falcon glass culture slides. Collagen 2(I) mRNA was detected using an antisense riboprobe synthesized with T3 RNA polymerase, using 35S-UTP (approximately 800 Ci/mmol; Amersham, Little Chalfont, UK), and plasmid was prepared from I.M.A.G.E. Consortium Clone ID 3415562 (35) linearized with XhoI. Antisense probe was used without hydrolysis. Luciferase mRNA was detected using an antisense riboprobe synthesized with T7 RNA polymerase, using 35S-UTP (approximately 800 Ci/mmol; Amersham) and BstE II–linearized pGEM-luc Vector (Promega, Madison, WI). The regions of sequence that were used to produce the luciferase riboprobes did not show significant homology to any known mouse sequences (http://www.ncbi.nlm.nih.gov/BLAST/).
The methods for pretreatment, hybridization, washing, and dipping of slides in Ilford K5 for autoradiography were as described by Senior et al. (36) for formalin-fixed paraffin-embedded tissue (37), with modifications. The presence of hybridizable mRNA in all compartments of the tissues studied was established in nearby sections using an antisense -actin probe.
Autoradiography was at 4°C (two exposures per probe at 10 and 18 d for collagen 2[I] and luciferase, one exposure at 10 d for -actin mRNA), before developing in Kodak D19 and counterstaining by Giemsa’s method. Sections were examined under conventional or reflected light/dark-field conditions (Nikon ME600; Nikon UK Ltd., Kingston upon Thames, UK) that allowed individual autoradiographic silver grains to be seen as bright objects on a dark background.
Combined Immunohistochemistry and ISH for the Y Chromosome
Standard immunohistochemical techniques were used to stain for -SMA. In brief, 4-µm-thick sections of formalin-fixed, paraffin-embedded tissue were rehydrated through xylene and graded alcohols. The sections were blocked for endogenous peroxidase (30% H2O2 diluted 1:20 in methanol, 10 min room temperature) and alkaline phosphatase (methanol:acetic acid 3:1 at 4°C for 1 min) activity. Sections were blocked in normal rabbit serum (1:25, X0902; Dako UK Ltd., Ely, UK). Monoclonal primary antibody mouse anti--SMA (1:4000, A-2547; Sigma, St. Louis, MO) was applied for 30 min. The sections were rinsed in PBS, then biotinylated rabbit anti-mouse antibody (1:300, E0354; Dako) was applied for 35 min. After rinsing again in PBS, streptavidin-alkaline phosphatase (1:50, D0396; Dako) was applied for 35 min, followed by further rinsing in PBS. Alkaline phosphatase was visualized using Vector Red (SK-5100; Vector Lab, Burlingame, CA).
After immunohistochemistry, sections were prepared for ISH. The slides first were permeabilized in sodium thiocyanate, then digested in pepsin 0.4%/0.1 M HCl at 37°C using several digestion times (1 to 15 min) to obtain optimal probe signals. Digestion was stopped by quenching in glycine 0.2% in 2 x PBS for 2 min. After rinsing in PBS, slides were postfixed in 4% paraformaldehyde for 2 min, rinsed in PBS, passed through graded alcohols, and then air-dried. A mouse Y chromosome probe (fluorescein-labeled, 1189-YMF-01; Cambio Ltd., Dry Drayton, UK) was diluted 1:250 in our own HybMix buffer (0.024% Denhardt’s [in 10x salts], 60% deionized formamide, 12% dextran sulfate, 3.5% ribosomal RNA, and 10 mM dithiothreitol). Approximately 20 µl of probe/HybMix solution was applied to the slides, which then were sealed with rubber cement and heated on a hot plate for 10 min at 60°C. The slides then were left overnight at 37°C in a moist chamber sealed with tape. Rubber cement was removed, and slides were washed twice with 0.5x SSC at 37°C for 5 min. The sections were visualized indirectly after application of a peroxidase-labeled anti-fluorescein antibody (1:250; Roche Products Ltd., Hertfordshire, UK), followed by a brief incubation with 0.5 mg/ml 3,3' diaminobenzidine in 30% hydrogen peroxide.
Negative controls for ISH included hybridizing with HybMix only and probing sections of female mouse kidney. Positive controls were performed on male transgenic mouse sections.
Statistical Analyses
Data are presented as mean ± SE. Two-tailed, paired t tests were calculated using GraphPad Prism (GraphPad Software Inc., San Diego, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
On light microscopy, at day 14 after UUO, there was interstitial expansion, with an increase in interstitial cells, including some inflammatory cells. Several tubules were dilated with a flattened epithelium, particularly in the cortex (Figure 1, A and B).
|
Increased collagen 2(I) mRNA levels were apparent at day 14 in the obstructed kidney by ISH using a radiolabeled riboprobe (Figure 2, A and B). The results were identical in all animals tested, whether wild-type, transgenic, or BM-transplanted. Collagen mRNA was present in the interstitial cells in a peritubular reticular pattern.
|
|
2(I) mRNA distribution (Figure 2, C and D). The luciferase probe gave no specific hybridization signals in control and obstructed kidneys of mice that received wild-type–to–wild-type BM transplant (Figure 2, E and F).
The transgenic mice also expressed -galactosidase as a reporter molecule. This reporter molecule was not useful in our study because it could not be distinguished from enhanced endogenous -galactosidase activity in the obstructed kidney (42).
BM Cells from Transgenic Mice Can Produce Collagen I and Luciferase In Vitro
Cell cultures that were enriched in mesenchymal stem cells were assessed for collagen I and luciferase expression. Mesenchymal BM cells that were grown on BD Falcon glass culture slides were analyzed by in situ hybridization for collagen I. Abundant transcripts were found (Figure 4A). Six-well-plate mesenchymal cell cultures were incubated with luciferin, and luciferase activity was assessed using IVIS Imaging System 100 Series photometer, which detects bioluminescence that is produced when luciferin is degraded by luciferase enzyme. Transgenic BM cell cultures were found to produce luciferase, whereas control wild-type BM cells that were cultured in the same conditions were negative (Figure 4, B and C). These results show that the transgene can be activated in the donor BM cells.
|
|
Absence of Luciferase in the Obstructed Kidneys of BM-Chimeric Mice
No luciferase activity was detected above background levels in tissue homogenates of control and obstructed kidneys in BM-chimeric mice at day 7 or 14 after UUO (Figure 3). ISH for luciferase transcripts showed signals in only two mice that received a transplant of the 14 analyzed. In one mouse, very few tubular cells were positive (Figure 6, A through D). In the other mouse, a single interstitial cell with luciferase mRNA was seen (Figure 6, E and F).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
In all of the above studies, the precise identity of the BM-derived cells remains to be proved. First, the use of immunohistochemistry to identify cell types is not entirely reliable in view of the close apposition of cells in the interstitial space. Cell processes may surround BM-derived inflammatory cells, giving rise to spurious results. Second, there is no universally sensitive and specific immunohistochemical marker for fibroblasts and myofibroblasts. SMA also stains smooth muscle cells and pericytes, and vimentin stains a variety of mesenchymal cell types. The use of FSP1 as a specific marker of fibroblasts also has been called into question (46,47).
We therefore decided to search for production of collagen I by the BM-derived cells, an approach that investigates whether there are any functional BM-derived fibroblasts or myofibroblasts. We used a line of transgenic mice in which production of collagen I was accompanied by increased expression of reporter genes, and first verified that UUO led to appropriate collagen and reporter molecule expression in the transgenic line. We then created a BM-chimeric mouse in which production of collagen I by BM-derived cells but not by intrinsic renal cells would lead to concomitant synthesis of reporter molecule luciferase in the obstructed kidney. Using enzyme assays for luciferase detection, we were unable to confirm participation of BM-derived cells in collagen I production. ISH for luciferase mRNA showed a trivial number of tubular cells and a single interstitial cell that contained transcripts. Both of these detection methods are sensitive and specific and were previously validated on transgenic animals, making a false-negative result unlikely. Experiments were carried out at two time points after UUO (days 7 and 14), allowing assessment of collagen production at different stages.
As collagen I is a major component of the ECM, we think that our study provides compelling evidence that indigenous kidney cells are more likely responsible for matrix increase in kidney fibrosis. In view of the results of previous studies in the kidney based on morphology alone, which suggested a large participation of BM-derived cells in kidney fibrosis, we think that our study highlights the necessity of using functional evaluation of BM-derived cells in addition to morphologic assessment. We were able to reproduce previous observations of BM-derived interstitial SMA-positive cells. As mentioned above, interpretation of combined immunohistochemistry with ISH for the Y chromosome is difficult in the confined interstitial space, but we are convinced that at least some of the Y-positive, SMA-positive cells are real. There are two possible explanations for the discrepancies between morphologic and functional studies. First, not all SMA-positive cells necessarily produce collagen (47); indeed, in the rat, most collagen-producing cells are SMA negative (7). Second, reporter molecule expression could be incomplete and not always correlated with collagen type I production, depending on the fibrogenic stimulus and the organ considered. Previous studies on our transgenic strain and our own positive control experiments in transgenic mice ruled this out.
In other organs, fibroblasts have been shown to differentiate from BM and implicated in collagen synthesis (22–25,48). Although a number of these studies rely on morphologic assessment only, we consider that it is possible that (myo)fibroblasts from different organs may have different origins, potentially recapitulating their embryologic development. This is supported by studies showing heterogeneity in fibroblast phenotype and function depending on the organ studied (49,50). The renal interstitium derives in the embryo from the metanephric mesenchyme and is divided into cortical and medullary compartments, which have distinct structural, developmental, and endocrine characteristics, such as the production of erythropoietin and adenosine (4). These functions are distinctly organ specific and may rely on a local source for regeneration.
|
Conclusion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
|
Acknowledgments |
|---|
Part of this work was presented as an abstract at Renal Week 2005 in Philadelphia, PA, November 8 to 13, 2005.
We acknowledge George Elia, Pooja Seedhar, and Toby Hunt (Histopathology Unit, London Research Institute, Cancer Research UK) for help with histology and ISH; Dave Phillips (Sequencing Unit, London Research Institute, Cancer Research UK); and Prof. Martin Bromley and the Imaging Sciences Department, MRC Clinical Sciences Centre, Imperial College, Hammersmith Campus.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. Strutz How Many Different Roads May a Cell Walk down in Order to Become a Fibroblast? J. Am. Soc. Nephrol., December 1, 2008; 19(12): 2246 - 2248. [Full Text] [PDF]
|
|
|
|
 |
|
 S.-L. Lin, T. Kisseleva, D. A. Brenner, and J. S. Duffield Pericytes and Perivascular Fibroblasts Are the Primary Source of Collagen-Producing Cells in Obstructive Fibrosis of the Kidney Am. J. Pathol., December 1, 2008; 173(6): 1617 - 1627. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Stokman, J. C. Leemans, I. Stroo, I. Hoedemaeker, N. Claessen, G. J. D. Teske, J. J. Weening, and S. Florquin Enhanced mobilization of bone marrow cells does not ameliorate renal fibrosis Nephrol. Dial. Transplant., February 1, 2008; 23(2): 483 - 491. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Li, J. A. Deane, N. V. Campanale, J. F. Bertram, and S. D. Ricardo The Contribution of Bone Marrow-Derived Cells to the Development of Renal Interstitial Fibrosis Stem Cells, March 1, 2007; 25(3): 697 - 706. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Poulsom, E. I. Prodromidi, C. D. Pusey, and H. T. Cook Cell therapy for renal regeneration--time for some joined-up thinking? Nephrol. Dial. Transplant., December 1, 2006; 21(12): 3349 - 3353. [Full Text] [PDF]
|
|
|
|
|
|
F. Strutz and M. Zeisberg Renal Fibroblasts and Myofibroblasts in Chronic Kidney Disease J. Am. Soc. Nephrol., November 1, 2006; 17(11): 2992 - 2998. [Full Text] [PDF]
|
|
|
|
|
|
E. I. Prodromidi, R. Poulsom, R. Jeffery, C. A. Roufosse, P. J. Pollard, C. D. Pusey, and H. T. Cook Bone Marrow-Derived Cells Contribute to Podocyte Regeneration and Amelioration of Renal Disease in a Mouse Model of Alport Syndrome Stem Cells, November 1, 2006; 24(11): 2448 - 2455. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
, 
); O, obstructed kidney (
, 
, 
); O7, day 7 after obstruction; O14, day 14 after obstruction. Results are expressed in light units per milligram of protein. Significant luciferase activity was detected only in transgenic animals, and levels were significantly increased with obstruction (P = 0.0013).
