Cisplatin-Induced Renal Interstitial Fibrosis in Neonatal Rats,Developing as Solitary Nephron Unit Lesions

Abstract

Cisplatin (CDDP)-induced renal lesions in rats prove a useful model for analysis of the pathogenesis of post-tubular injury-renal interstitial fibrosis. This study investigated the histopathological changes in 10-day-old neonatal rats induced by a single injection of CDDP (4.5 mg/kg). Compared with age-matched controls, on postinjection (PI) days 1 to 6, the number of apoptotic cells, demonstrable with TUNEL method, was significantly increased in CDDP-treated neonates, and there was no marked epithelial necrosis nor fibrotic lesions. Fibrotic lesions began to be developed solitarily around some nephrons with dilated ducts in the corticomedullary junction on PI day 10 and the lesions became more prominent until PI day 20. The α-SMA-positive myofibroblastic cells were seen exclusively in the fibrotic lesions. Additionally, the numbers of macrophages reacting with ED1 (specific for exudate macrophages), ED2 (for resident macrophages), and OX6 (recognizing MHC class II antigens expressed in antigen-presenting macrophages/dendritic cells) were significantly increased around the affected renal tubules. A greater immunoreaction for TGF-β1 was seen mostly in the renal epithelial cells of CDDP-treated neonates. These findings indicated that macrophage populations and myofibrolastic cells as well as TGF-β1 may be responsible for the production of neonatal renal interstitial fibrosis. Compared with CDDP-injected adult rats that develop extensive interstitial fibrosis (Yamate et al., J Comp Pathol, 1995), the formation of fibrotic lesions was delayed, and the lesions were limited to the area around the affected nephrons; this could be attributable to differences in renal morphology between neonates and mature kidney of adult rats.

Introduction

Progressive interstitial fibrosis is the key determinant of end-stage renal disease, regardless of the primary disease process (Cheng and Lovett, 2003). Thus, interstitial fibrosis has a greater impact on the progression of chronic renal disease than glomerulosclerosis (Jones et al., 1992; Mackensen-Haen et al., 1992). The fibrotic lesions are evoked through complicated processes after tissue injury, requiring the participation of various inflammatory cells and abnormal accumulation of ECMs (Kuncio et al., 1991; Tamaki et al., 1994). In particular, monocyte-derived macrophages responding to tissue injury play a pivotal role in fibrosis via production of growth factors such as TGF-β1 (Johnson et al., 1992; Tamaki et al., 1994; Jones et al., 1999; Frazier et al., 2000; Yamate et al., 2001). TGF-β1 mediates the development of myofibroblastic cells capable of producing ECMs such as collagens and fibronection, leading to the formation of fibrotic lesions (Alvarez et al., 1992; Creely et al., 1992).

To clarify the complicated mechanisms of renal fibrosis, nephrotoxic chemicals such as adriamycin (Tamaki et al., 1994), puromycin aminonucleoside (Diamond et al., 1989; Jones et al., 1992; Eddy, 1996), cyclosporin (Young et al., 1995), gentamicin (Geleilete et al., 2002) and mercuric chloride (Suzuki et al., 1999) have been used for the purpose of inducing renal lesions in rats. In addition to these models, long-term unilateral ureteral obstruction (Diamond et al., 1995; Yamate et al., 1998) and 5/6 nephrectomy (Ng et al., 1998) are other useful models for the fibrosis. We have previously shown that CDDP, an anticancer drug with nephrotoxicity, induced renal interstitial fibrosis in rats (Yamate et al., 1995). The initial lesion is characterized by extensive tubular damage of the lower straight part (P₃ segment) of the proximal tubule in the corticomedullary junction. CDDP-induced renal lesions should prove a useful model of post-tubular injury-interstitial fibrosis.

Recently, it has been reported that the fetus has the ability to heal skin injury without fibrosis and subsequent scar formation (Cass et al., 1997), suggesting the differences in mechanisms of fibrosis depending on age. The aforementioned experimental models for renal fibrosis have been developed using adult rats more than 6 weeks old. To our knowledge, there have been no experiments using fetal or neonatal rat kidneys. In rats, maturation of kidneys is completed up to 15 days after birth (Coles et al., 1993; Chevalier, 1996; Nagata et al., 1996; Basile and Hammerman, 1998; Liu et al., 1999); a balance between cellular proliferation and apoptosis, as well as growth factors and adhesion molecules are thought to contribute to renal morphogenesis (Coles et al., 1993; Chevalier, 1996; Basile and Hammerman, 1998; Liu et al., 1999). In order to better understand the basis of renal interstitial fibrosis, we examined pathological changes that develop in CDDP-injected neonatal rats with special reference to apoptosis, cellular proliferation, macrophage participation, myofibroblastic cell development and TGF-β1 expression. Interestingly, renal interstitial fibrosis in neonatal rats treated with CDDP developed as separated nephron unit lesions, differing from CDDP-treated adult rats showing extensive renal interstitial fibrosis (Yamate et al., 1995).

Materials and Methods

CDDP Injection

The following experiments were in compliance with our institutional guidelines for animal care. Forty-two neonatal rats of F344 strain (Charles River Japan, Hino, Shiga, Japan) at the age of 10 days, weighing 12–18 g, were used. They were housed in an animal room controlled at 22 ± 3°C and with a 12 h light-dark cycle, being allowed free access to a standard commercial diet (MF, Oriental Yeast Co. Ltd., Tokyo, Japan) and tap water. CDDP (10 mg/20 mL in a vial; Nippon Kayaku Co., Ltd, Tokyo, Japan) was injected ip into 24 rats at a single dose of 4.5 mg/kg body weight; the dose was capable of inducing sublethal renal failure in neonates without mortality, according to a preliminary experiment. Four rats were euthanized on days 1, 3, 6, 10, 15, and 20 after the CDDP dosing (PI day). The remaining 18 rats were ip injected with PBS; 3 rats were euthanized on the same days as described, and served as controls. Because neonatal kidneys have nephrogenesis and thus the morphology may change during development, at each examination point controls were put and compared with CDDP-treated rats, respectively.

One hour prior to euthanasia, all animals received an ip injection of BrdU (10 mg/mL in PBS), 50 mg/kg body weight. Serum samples from blood obtained under ether anesthesia were tested for BUN and creatinine with a Clinical Analyzer 7170 (Hitachi, Tokyo, Japan).

Histopathology and Immunohistochemistry

Kidney was divided into 2 portions by cross-cut. In each animal, half of 1 kidney was fixed in 10% neutral buffered formalin, and the remaining half was used for RT-PCR analysis; half of the other kidney was fixed in methacarn solution and the remainder in Zamboni’s fixative (0.21% picric acid and 2% paraformaldehyde in 130 mM phosphate buffer, pH 7.4) (Ide et al., 2003). These renal tissues were embedded in paraffin, and sectioned at 3–4 μm in thickness. Formalin-fixed, deparaffinized sections were stained with HE and azan-Mallory for fibrotic lesions.

Immunohistochemical staining was performed by the Avidin-Biotin Complex method (LSAB kit; Dako Corp., Carpinteria, CA) with primary antibodies listed in Table 1. The methods have been described elsewhere (Yamate et al., 2002a; Ide et al., 2003). Briefly, after pretreatments applied for each primary antibody (Table 1) to inactivate endogenous peroxidase, the deparaffinized sections were treated with 5% skimmed milk for 30 minutes, and incubated with each primary antibody for 14–24 hours. For BrdU staining, before the pretreatment, sections were treated with 4N HCl at 37°C for 20 minutes to denature DNA. Incubation with biotinylated goat anti-mouse antibody for monoclonal antibodies or goat anti-rabbit antibody for polyclonal antibodies followed. Final incubation was carried out for 1 hour with an avidin-horseradish peroxidase complex, and positive reactions were visualized with DAB. Sections were lightly counterstained with hematoxylin. Nonimmunized mouse or rabbit serum, used instead of primary antibodies, served as negative controls.

TUNEL for Apoptotic Cells

A standard in-situ TUNEL (Apop Tag Kit; Oncor Inc., MD) method was used for detection of DNA fragmentation in apoptotic cells (Lan et al., 1997). After treatment with proteinase K for 20 minutes at 37°C and then with 3% hydrogen peroxide for 20 minutes at room temperature to quench endogenous peroxidase activity, formalin-fixed, deparaffinized sections were incubated with TdT and digoxigenin DNA labeling mixture in the TdT reaction buffer for 60 minutes at 37°C. Subsequently, the sections were blocked with 10% normal goat serum and 10% fetal calf serum for 30 minutes at 37°C, and then treated with horseradish peroxidase-conjugated anti-digoxigenin for 60 minutes. Peroxidase was detected with DAB. The sections were counterstained lightly with hematoxylin. In negative controls, TdT enzyme was replaced with distilled water.

RT-PCR for TGF-β1 mRNA Expression

Total RNA was extracted from fresh renal tissues using Trizol Reagent (Invitrogen Corp., Carlsbed, CA; Chomczynski, 1993). The RNA was reverse-transcribed to cDNA using the Super Script Preamplification System (Invitrogen Corp.). cDNA was amplified by PCR with Taq DNA polymerase (TaKaRa Shuzo, Otsu, Japan) and using specific primers for rat TGF-β1 or β-actin (control). The following conditions were used for the amplification: for TGF-β 1, 25 cycles of 1 minute of denaturation at 94°C, 1 minute of annealing at 60°C, and 1 minute of synthesis at 72°C (Tanuma et al., 1997). The oligonucleotides used were as follows: for TGF-β, sense 5′-CTTCAGCTCCACAGAGAAGAACTGC-3′and antisense 5′-CACGATCATGTTGGACAACTGCTCC-3′ (Tanuma et al., 1997): for β-actin, sense 5′-TAAAGACCTCTATGCCAACAC-3’ and antisense 5′-CTCCTGCTTGCTGATCCACAT-3′ (Tsujino et al., 1997). The PCR products were subjected to electrophoresis in a 1% agarose gel. DNA was stained with ethidium bromide on the gel. The bands were semiquantitatively evaluated with image analysis software (NIH Image 1.61, Bethesda, MD) (Tsujino et al., 1997).

Evaluation and Statistical Analysis

Cells showing a distinct positive-reaction for ED1, ED2, OX6, α-SMA, and BrdU immunohistochemistry, as well as TUNEL method-positive cells were counted in 5 randomly selected areas (0.0625 mm²/area) in the corticomedullary junction of cross-sections at a magnification of X400; the counts were done without knowledge of the treated or control group. Data obtained in the immunohistochemical examinations, TUNEL method-positive cells, body weight changes, and blood biochemistry values are presented as mean ± standard error (SE). Paired samples between control and CDDP groups at each examination point were compared by Student’s t-test and values of p < 0.05 were considered significant. The cell count and evaluation methods have been described in our previous papers (Yamate et al., 1995, 1998, 2000a, 2002b).

The reactivities for TGF-β1 immunohistochemistry were assessed semiquantitatively in the renal tubules and macrophages in the corticomedullary junction with fibrotic areas (Diamond et al., 1995): ±, less faintly positive staining; +, faintly positive staining; 2+, moderately positive staining; 3+, strongly positive staining.

Results

CDDP-Induced Renal Lesions

As shown in Figure 1, the body weight of control rats gradually increased up to PI day 20, whereas that of CDDP group was significantly suppressed between PI days 3 and 20. BUN (45 ± 10 mg/dl) and creatinine (0.7 ± 0.2 mg/dl) values in the CDDP group tended to increase on PI day 6 in contrast to controls (BUN, 32 ± 9 mg/dl; creatinine, 0.30 ± 0.25 mg/dl), but no statistically significant difference was noted. Mean values of BUN and creatinine on other PI days in both control and CDDP groups were within normal ranges, between 15 and 32 mg/dl or between 0.1 and 0.3 mg/dl, respectively.

In HE-stained sections, no histopathological changes were observed in all control rats and rats of CDDP group on PI days 1 to 6. Immature glomeruli and renal tubules were still seen in kidneys of both control and CDDP rats on PI day 1 (Figure 2). These structures disappeared by PI day 6, indicating completion of mature kidneys onwards. On PI day 10, in CDDP-injected rats, there were some nephrons with variously dilated ducts in the corticomedullary junction; epithelial cells in the affected renal tubules had swollen cytoplasm and an enlarged nucleus, and were occasionally desquamated into tubular lumina. On PI days 15 and 20, dilation of the affected tubules was more prominent, being accompanied with infiltration of mononuclear cells around these tubules and within their lumina, and some of these tubules were lined by regenerating, flattened epithelial cells with basophilic cytoplasm (Figure 3). On PI day 10, fibrotic areas consisting of spindle-shaped fibroblastic cells began to be seen around the affected tubules, and on PI days 15 and 20, the fibrotic areas, stained blue by azan Mallory stain, became more evident around the dilated tubules (Figure 4), but extensive fibrosis was not observed. In the fibrotic areas around the affected tubules, α-SMA-positive myofibroblastic cells were seen (Figure 5). In keeping with the formation of fibrotic areas on PI days 10 to 20, the number of α-SMA-positive myofibroblastic cell number was significantly increased (Figure 6).

Proliferating Cells and Apoptotic Cells

As shown in Figure 7, BrdU-positive proliferating cells (S phase) were frequently seen on PI day 1 in both control and CDDP rats; the number gradually decreased until PI day 20. Statistically, BrdU-positive cell number on PI day 6 was significantly decreased in the CDDP group. In the control group, TUNEL-positive apoptotic cells were often seen on PI days 1 to 6, but the cells were rarely seen on PI days 10 to 20 (Figure 8). The TUNEL-positive cell number in the CDDP group was also more frequently seen on PI days 1 to 6 than 10 to 20; the number on PI days 3, 6, 10, 15, and 20 showed significant increase as opposed to that of controls (Figure 8). BrdU- and TUNEL-positive cells are shown in Figures 9 and 10, respectively.

Macrophage Populations

Cells reacting to ED1 (Figure 11), ED2 (Figure 12), and OX6 (Figure 13) were diffusely seen in control kidneys; intriguingly, ED1- and ED2-positive cells were more frequently seen on PI days 1 to 6, but the numbers tended to decrease onwards. Statistically, the number of ED1-positive cells on PI days 1, 6, 10, 15, and 20 was significantly greater in the CDDP group than in control group (Figure 11). The ED2-positive cell number in the CDDP group showed a significant increase on PI days 1, 10, 15, and 20, in contrast to that in the control group (Figure 12). The number of OX6-positive cells on PI days 10, 15, and 20 was significantly increased in the CDDP group compared to controls (Figure 13). In the CDDP group, ED1- and ED2-positive cells on PI days 10 to 20 were present mainly around the dilated tubules (Figures 14 and 15). Although OX6-positive cells were sporadically seen in the interstitium of control kidneys (Figure 16), the positive cells on PI days 10 to 20 in the CDDP group were localized exclusively around the affected tubules (Figure 17).

TGF-β1 Immunoexpression and RT-PCR Analysis for TGF-β1 mRNA

As shown in Table 2, strongly positive staining for TGF-β1 was seen in the renal tubules of the control group on PI day 1, and the reactivity was gradually decreased on PI days 3 and 6 (2+), and on PI days 10 to 20 (+) (Figure 18). In the control group, there was no immunoexpression for TGF-β1 in macrophages. In contrast, the moderately (2+) and markedly (3+) positive staining patterns were consistently seen in renal tubules of CDDP group on PI days 1 to 20 (Figure 19); especially, epithelial cells around the dilated renal tubules and within fibrotic areas showed a greater expression. Interestingly, macrophages reacting to TGF-β1 were occasionally seen in the CDDP group on PI days 10 to 20 (Inset in Figure 19), but the reactivity was less faintly positive (Table 2).

No marked difference in TGF-β1mRNA expression by RT-PCR methods was noted between control and CDDP groups on each PI day (Figure 20).

Discussion

CDDP has a toxic effect on the lower straight part (P₃ segment) of the proximal tubules of rats (Jones et al., 1985; Yamate et al., 1995; Lieberthal et al., 1996). We have previously shown that progressive interstitial fibrosis develops in CDDP-injected adult rats at the age of 6 to 7 weeks (Yamate et al., 1995, 2000a). The CDDP-induced primary lesions were characterized by nuclear alteration, necrosis and desquamation in epithelial cells of proximal renal tubules in the corticomedullary junction; such lesions were observed seen as early as 1 day after a single injection of CDDP (6 mg/kg body weight). These changes became more prominent on PI days 3 to 5. On PI day 7, regenerating renal epithelial cells replaced the injured epithelial cells, and thereafter, the fibrosis progressed around variously dilated renal tubules in the corticomedullary junction. On PI days 10 to 28, the fibrotic areas were formed extensively in the corticomedullary junction of CDDP-treated adult rats.

In the present study, markedly suppressed bodyweight gain of CDDP group indicated toxicological effects of CDDP on neonatal rats. Recently, it has been reported that apoptosis may be induced in renal epithelial cells by CDDP (Lieberthal et al., 1996; Yamate et al., 2000a). The TUNEL-positive cell number was significantly increased on PI days 3, 6, 10, 15, and 20 in CDDP group, indicating that CDDP might induce apoptosis in renal epithelial cells of rats. In HE-stained sections, the apoptotic cells appeared to have segregated or condensed hyperchromatin. Besides apoptosis, necrosis, characterized morphologically by swollen cytoplasm and enlarged nuclei, were other changes observed with CDDP-induced cell damage in the renal epithelial cells (Yamate et al., 1995, 2000a; Lieberthal et al., 1996). In this study, epithelial necrotic changes were not clearly seen in the corticomedullary junction of CDDP groups on PI days 1 to 6.

Rat kidneys are completely mature around 15 days after birth (Chevalier, 1996). In normal nephrogenesis, apoptosis and cellular proliferation play crucial roles (Haralambous-Gasser et al., 1993; Chevalier, 1996; Nagata et al., 1996), and the maintenace of a balance between cell death and cell survival seems to determine the morphology of the kidneys (Nagata et al., 1996). Interestingly, up to PI day 6 (16 days old), TUNEL-positive epithelial cells were often seen even in the present control group, whereas such cells were rarely seen on PI days 10 to 20 in the control group. Furthermore, BrdU-positive cells were also seen frequently on PI days 1 to 6 in both control and CDDP groups. The frequent appearance of proliferating and apoptotic cells in the controls on PI days 1 to 6 might account for nephrogenesis (Coles et al., 1993; Chevalier, 1996). For CDDP group, it was hypothesized that high proliferative activity during tubulogenesis by PI day 6 might have partly compensated for the excessive loss of epithelial cells due to CDDP-induced apoptosis. Therefore, in CDDP-treated neonates, histopathological changes such as necrosis and desquamation of epithelial cells, as well as subsequent dilated duct formation, failed to develop up to PI day 6. Such histopathological findings have been observed as initial lesions in CDDP-treated adult rats on PI days 1 to 7 (Yamate et al., 1995, 2000a).

In CDDP-treated adult rats, α-SMA-positive myofibroblastic cells were observed in the interstitium of damaged corticomedullary junction stating on PI days 4 to 7, and the number gradually increased, resulting in the production of extensive fibrotic lesions (Yamate et al., 1995, 2000a). In the present study, such cell types began to be seen with a significantly increased number on PI days 10 to 20, although their distribution was almost limited to the area around some nephrons with dilated tubules. Extensive interstitial fibrosis did not develop until PI day 20 in the CDDP-treated neonates. These findings indicate that there were differences in the extent and time of onset of fibrotic lesions between CDDP-treated neonatal and adult rats. As mentioned above, in the developing kidney, many renal tubules could mature without epithelial damage due to CDDP injection through an appropriate balance between apoptosis and cellular proliferation. However, we hypothesize that some nephrons can not withstand damage caused by CDDP, resulting in epithelial cell injury (desquamation) and subsequent formation of dilated ducts rimmed by regenerating flattened epithelial cells. Subsequently, fibrotic lesions develop around the nephrons with dilated ducts as solitary lesions. On PI day 6, the significantly decreased number of BrdU-positive proliferating cells in the CDDP group suggests the following explanation for the pathogenesis of the aberrant tubulogenesis; because of an increased number of apoptotic cells, an insufficient number of epithelial cells was present in some nephrons, resulting in dilated duct formation. This phenomenon appears to be similar to that observed in renal hypoplasia in bcl-2-deficient mice where the lesion is induced by the occurrence of widespread apoptosis during nephrogenesis (Nagata et al., 1996). The tendency in BUN and creatinine values to increase on day 6 might be explained by the presence of somewhat abnormal tubulogenesis in some nephrons.

Perhaps, more interestingly, the present study revealed that different macrophage populations appeared around the affected renal tubules in the CDDP group. These macrophage populations were identified by immunohistochemistry with ED1, ED2 and OX6. Previously, we reported that ED1-positive macrophages took part in the early development of CDDP-induced fibrotic lesions (Yamate et al., 1995). ED1 has been used widely to detect infiltrating macrophages in experimentally induced rat pathological lesions, and it is a monoclonal antibody that labels monocytes and activated macrophages, particularly exudate macrophages (Dijkstra et al., 1985; Tamaki et al., 1994; Desmouliere et al., 1995; Diamond et al., 1995; Nakatsuji et al., 1997; Cook et al., 2002). Antigens recognized by ED1 are present on the membrane of cytoplasmic granules, especially phagolysosomes, of macrophages, and the degree of ED1 expression depends on phagocytic activity (Dijkstra et al., 1985; Damoiseaux et al., 1994). Significantly increased number of ED1-positive cells on PI days 1 and 6 in the CDDP group might be responsible for removal and clearance of apoptotic epithelial cells (Desmouliere et al., 1995; Lan et al., 1997). And, frequent appearance of the ED1-positive cells in the controls on PI days 1 to 6 might be also explained by occurrence of apoptosis relating to normal nephrogenesis (Coles et al., 1993; Chevalier, 1996). Additionally, macrophages are thought to play a role in the induction of the myofibroblastic cells in experimental rat renal fibrosis in response to tissue injury via the production of fibrogenic factors such as TGF-β1 (Jones et al., 1992; Tamaki et al., 1994; Diamond et al., 1995; Akagi et al., 1996; Border and Noble, 1997; Yamate et al., 2001). On PI days 10 to 20, increased numbers of ED1-positive cells around the affected tubules in the CDDP group might contribute to the development of myofibroblastic cells, because α-SMA-positive myofibroblastic cells began to appear simultaneously in the same areas. In fact, a small number of macrophages reacting to TGF-β1 were seen in the CDDP group on PI days 10 to 20 (Table 2).

In addition to exudate macrophages, macrophage populations include resident macrophages (so-called histiocytes and Kupffer cells in the liver) and cells differentiating into dendritic cells (Takahashi et al., 1996; Rescigno et al., 1997; Yamate et al., 2000b). ED2 has been widely used to detect resident macrophages in rat lesions (Johnson et al., 1992; Hines et al., 1993; Nakatsuji et al., 1997). In the present study, although the number was very small, the ED2-positive cells appeared around the dilated tubules, with a significantly-increased number on PI days 10 to 20. In CCl₄-and cholestatic-induced rat liver fibrosis, ED1- and ED2-positive cells were frequently seen as major cell types; their appearance was correlated to the development of myofibroblastic cells (Johnson et al., 1992; Hines et al., 1993). These macrophage populations have been considered to have phagocytic activity and to produce fibrogenic factors capable of inducing myofibroblastic cells in the injured lesions (Moriyama et al., 1997; Friedman, 2000). However, the biological functions of antigens recognized by ED2 have not yet been determined (Dijkstra et al., 1985; Damoiseaux et al., 1994). Previously, we have reported in experiments using adult rats that no ED2-positive cells were detected in ureteral obstruction-induced and CDDP-induced renal fibrosis (Yamate et al., 1995, 1998). Although the reason why there is a difference in appearance of ED2-positive cells in fibrotic lesions between neonatal and adult rats is unclear, the distribution of resident macrophages in the developing kidney may be different from that in the mature kidney; because, unexpectedly ED2-positive cells were often seen on PI days 1 to 6 in the controls.

OX6 is a monoclonal antibody against rat MHC class II antigen which recognizes antigen-presenting cells such as stimulated macrophages and dendritic cells (Zhang et al., 1993; Yamashiro et al., 1994; Yamate et al., 2002). The present study showed for the first time that OX6-positive cells participated in the renal fibrotic lesions; the number was significantly increased on PI days 10 to 20 in the CDDP group, in agreement with an increase in ED1- and ED2-positive cell numbers. Dendritic cells are “professional” antigen-presenting cells and their MHC class II expression is accelerated by inflammatory stimuli (Rescigno et al., 1997; Fields et al., 2003). MHC class II-expressing cells present antigens to unprimed T cells, and subsequently Th1/Th2 lymphocyte commitment is determined (Rescigno et al., 1997; Thompson et al., 1998). However, in the present CDDP-treated neonatal rats, no lymphocytes were observed around the affected renal tubules. Further study is needed to elucidate the roles of OX6-positive cells and the relationship with lymphocytes.

TGF-β1 is regarded as the most important fibrogenic factor contributing to induction of myofibroblastic cells in injured tissues (Jones et al., 1992; Tamaki et al., 1994; Diamond et al., 1995; Akagi et al., 1996; Border and Noble, 1997; Coker et al., 1997; Hellerbrand et al., 1999; Yamate et al., 2001). In addition, TGF-β1 has been shown to be produced in kidneys during their development. TGF-β1 blocks or initiates cellular proliferation or differentiation, presumably depending on cell type and the stage of development (Basile and Hammerman, 1998; Clark and Coker, 1998). The biological actions of TGF-β1 are very diverse. In the present study, on PI days 1 and 3, immunohistochemical expression for TGF-β1 in renal tubules was moderate (2+) or marked (3+) in both control and CDDP groups; the expression might be related to nephrogenesis (Basile and Hammerman, 1998; Liu et al., 1999). Interestingly, the expression was decreased (+) on PI days 10 to 20 in control rats with mature kidneys, whereas it remained high (2+ or 3+) in the CDDP group. In particular, epithelial cells around the dilated renal tubules and within fibrotic areas showed a greater expression.

It seems likely that the expression on PI days 10 to 20 in the CDDP group might have been responsible for the development of myofibroblastic cells. In ureteral obstruction- or adriamycin-induced renal interstitial fibrosis of adult rats, the immunoreaction for TGF-β1 also was localized in renal epithelial cells as well as in a limited number of macrophages (Tamaki et al., 1994; Wright et al., 1996; Yamate et al., 1998). Functional roles of TGF-β1 produced by renal epithelial cells might be different between developing kidneys and kidneys with fibrotic lesions (Basile and Hammerman, 1998; Liu et al., 1999). There was no marked difference in TGF-β1 mRNA expression between control and CDDP groups. Although the exact protein levels of TGF-β1 remain to be analyzed, TGF-β1 mRNA may be consistently expressed during nephrogenesis and the maintenance of matured kidneys or in response to renal injury (Akagi et al., 1996; Basile and Hammerman, 1998; Jones et al., 1999). Besides TGF-β1, other factors such as PDGF and NGF are mediators for development of fibrotic lesions (Micera et al., 2001; Pontén et al., 2003). Participation of such factors should be investigated further in neonatal renal fibrosis.

In conclusion, the present study has described differences in CDDP-induced renal fibrosis in neonatal rats compared to adult rats, with regard to the extent and time of onset of the lesions. In CDDP-treated neonatal rats, the formation of fibrotic lesions was delayed and limited to the area around some nephrons with dilated ducts. Differences between fetal and adult wound repair have been described, but the key mechanisms that mediate scarless fetal repair remain unclear (Cass et al., 1997). Presumably, cellular functions evoked by cell-cell or cell-matrix interactions via growth factors and adhesion molecules might be different, depending on age, which likely accounts in part for the age-dependent differences in wound healing (Cass et al., 1997; Wu et al., 1999). In addition, the present study showed the participation of different macrophage populations, identified by ED1, ED2, and OX6, in renal interstitial fibrosis. Recently, it was reported that macrophages infiltrating rat inflamed glomeruli showed heterogeneous functions and reactions (Minto et al., 2003). Macrophages have functions such as phagocytosis, antigen presentation and regulation of inflammatory cells (Rescigno et al., 1997; Takahashi et al., 1996; Yamate et al., 2000b). Since expression of these functions may be dependent on microenvironmental conditions or age (Cass et al., 1997; Chevalier, 1996; Minto et al., 2003), the relationship between macrophage populations and renal interstitial fibrosis should be investigated further.