2008-AJN-February-Liu

Original Report: Laboratory Investigation American

Journal of

Nephrology

Am J Nephrol 2008;28:555–568 DOI: 10.1159/000115290

Received: November 14, 2007 Accepted: December 14, 2007 Published online: February 1, 2008

Low-Dose Local Kidney Irradiation Inhibits Progression of Experimental Crescentic Nephritis

by Promoting Apoptosis Diange Liu a, b Arifa Nazneen a Takashi Taguchi a M. Shawkat Razzaque a, c a

Department of Pathology, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan; Institute of Nephrology, Zhong Da Hospital, Southeast University School of Medicine, Nanjing, China; c Department of Developmental Biology, Harvard School of Dental Medicine, Boston, Mass., USA b

Key Words Crescentic glomerulonephritis ⴢ Radiation treatment ⴢ Apoptosis ⴢ Caspase 3 ⴢ Caspase 7 ⴢ p53 ⴢ TUNEL ⴢ Immunohistochemistry ⴢ Western blotting

Abstract Background: Crescentic glomerulonephritis is a rapidly progressive form of nephritis and is usually resistant to therapeutic intervention. Apoptosis plays a role in the resolution of glomerulonephritis. We investigated the effects of local kidney irradiation on the progression of experimental crescentic glomerulonephritis. Methods: The following three experimental rat groups were generated: (1) Group I, shamoperated control (n = 12); (2) Group II, rats injected intravenously with rabbit anti-rat GBM antibody (nephrotoxic serum, NTS) (n = 23), and (3) Group III, a single low-dose irradiation of 0.5 Gy X-ray to both kidneys at days 6, 13, 20, and 27 after NTS injection (n = 55). Renal function and blood leukocyte count were examined in different groups of rats at various time points. Kidneys obtained at various time points were analyzed to determine the effects of radiation in experimental nephritis. Results: Radiation of the kidneys reduced the levels of blood urea nitrogen and serum creatinine compared with Group II nephritic rats of similar age (p ! 0.05 or 0.001). No apparent changes in blood leukocyte counts were noted in various experimental groups. Glomerular hypercellularity, crescents, global sclerosis and tubulointerstitial damage developed gradually in Group II rats,

© 2008 S. Karger AG, Basel 0250–8095/08/0284–0555$24.50/0 Fax +41 61 306 12 34 E-Mail karger@karger.ch www.karger.com

Accessible online at: www.karger.com/ajn

but were decreased (p ! 0.05 or 0.001) after radiation treatment. The extent of tubulointerstitial damage was also reduced, and radiation-associated histological improvements were accompanied by reduced infiltration of macrophages in the glomeruli and interstitium. The numbers of PCNA- and ED1-positive cells were reduced in the kidneys at 1 day postirradiation, of rats irradiated at 6 and 13 days after NTS injection, compared with Group II at similar time intervals (p !

0.05). A larger numbers of TUNEL-positive cells were noted at 1 day post-irradiation in rats irradiated at 6 and 13 days after NTS injection, compared with Group II at similar time intervals (p ! 0.05). Immunostaining for macrophages ED1 and TUNEL staining of serial sections of irradiated nephritic kidneys showed few ED1-positive macrophages stained for TUNEL. Overexpression of active caspases 3 and 7 was noted in irradiated kidneys, compared with the corresponding Group II rats at similar time intervals. Western blot analysis showed marked increase in active caspase 3 and active caspase 7 expression in irradiated kidneys compared with NTS injection only. A marked increase in the expression of p53 protein, which is closely related to radiation-induced apoptosis, was also observed in irradiated kidneys compared with NTS injection only. Conclusion: Our study showed that renal radiation can alter acute glomerular inflammation by inducing apoptosis of intrinsic and infiltrating cells in the kidney in a rat model of crescentic glomerulonephritis. Lowdose kidney irradiation can inhibit the progression of experimental nephritis through inducing apoptosis.

Copyright © 2008 S. Karger AG, Basel

Takashi Taguchi, MD, PhD Department of Pathology Nagasaki University Graduate School of Biomedical Sciences, 1-12-4, Sakamoto Nagasaki 852-8523 (Japan) Tel. +81 95 819 7053, Fax +81 95 819 7056, E-Mail taguchi@nagasaki-u.ac.jp

Introduction

The renal transport system is essential for the physiologic regulation of organic and inorganic ion balance [1–4]. Various systemic, immunological, and metabolic diseases can alter renal structural integrity to affect transport system [1–4]. Some of the diseases (i.e., systemic hypertension, mild diabetes) or processes (i.e., ageing) take a relatively longer time to reduce renal function [5–8], while some of the diseases, like crescentic glomerulonephritis, can dramatically reduce renal function in a very short period of time.

Crescentic glomerulonephritis is a rapidly progressive renal disease characterized by diffuse crescent formation and rapid deterioration of renal function. In crescentic glomerulonephritis, monocytes/macrophages and T lymphocytes accumulate in Bowman’s space and are considered to play a role in crescent formation [9]. Extensive crescent formation is accompanied by interstitial infiltration of these cells, and the subsequent accumulation of extracellular matrix correlates well with the outcome [10]. Crescentic nephritis occurs in a variety of human autoimmune disorders. Treatment with plasma exchange and immunosuppressive drugs is generally effective in patients treated sufficiently early, but carries a high risk of serious adverse effects [11].

A human study described local irradiation of the graft combined with conventional therapy in 2 renal transplant patients complicated with crescentic glomerulonephritis [12]. The results showed that radiation therapy resulted in reduction of urinary protein excretion, suggesting the potential advantage of radiation therapy for crescentic glomerulonephritis [12]. A novel scoring system has been described to predict the outcome of patients treated with radiation therapy for acute renal graft rejection [13]. However, the mechanism of the beneficial effect of irradiation remains obscure.

The mechanisms that control the progression of glomerulonephritis are also poorly understood, but recent studies indicate that apoptosis plays an important role in the resolution of intra- and extraglomerular inflammation and in the elimination of glomerular cells within the scaring regions for progressive crescentic glomerulonephritis [14, 15]. Apoptosis or programmed cell death is a distinct mode of cell destruction and represents a major regulatory mechanism in eliminating abundant and unwanted cells during embryonic development, growth, differentiation and normal cell turnover and anti-apoptosis therapy presents an option to restore pathological cellular imbalance [16–18]. Apoptosis is morphologically 556

Am J Nephrol 2008;28:555–568

characterized by condensation of nuclear chromatin, by blebbing of the nuclear and cytoplasmic membranes, and finally by fragmentation of the nuclear structures, leading to the formation of membrane-bound apoptotic bodies [16]. Increasing levels of macrophage proliferation and apoptosis were evident as crescents developed from a cellular to fibrocellular phenotype, with a dramatic reduction in both of these processes during the progression to a fibrotic phenotype, suggesting an important role for macrophages and apoptosis in the resolution of fibrocellular crescents to an acellular fibrotic structure [19].

Radiation-induced apoptosis has been a topic of intense research during the last decade. Radiation-induced apoptotic signaling can be initiated in different cellular compartments, including the nucleus, cytosolic elements and plasma membrane. There is general agreement that the tumor suppressor protein p53 plays a pivotal role in determining whether radiation-damaged cells will undergo apoptosis [20, 21]. Radiation-induced p53 activation causes a delay in cell cycle progression, predominantly at the G1-S transition, allowing the damaged DNA to be repaired before the occurrence of replication and mitosis [22–24]. On the other hand, the family of proteases, called caspases, is the key effector of cellular death [25–27]. Caspases can be broadly classified into initiator and effector caspases. Initiator caspases (e.g., caspases 8, 9 and 10) act upstream in the protease hierarchy and activate effector caspases (e.g., caspases 3, 6 and 7), which are considered as executioners of the apoptotic program.

In the present study, we investigated whether local radiation treatment can prevent the progression of crescentic glomerulonephritis by promoting apoptosis and the underlying regulatory mechanism, using the terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end labeling (TUNEL) method, immunohistochemistry and Western blotting. Our results showed that local radiation treatment can inhibit the progression of rat crescentic glomerulonephritis by radiation-induced apoptosis, and that caspases 3 and 7 are the key mediators of such apoptosis.

Materials and Methods Experimental Design Inbred male Sprague-Dawley rats (body weight 150–200 g, n = 90) were obtained from SLC Inc., Japan. They were divided into three experimental groups. Accelerated anti-GBM glomerulonephritis was induced and radiation treatment was performed as described previously [28, 29]. Group I (n = 12, control) con-

Liu /Nazneen /Taguchi /Razzaque

sisted of sham-operated control rats that received no radiation treatment or induction of anti-GBM glomerulonephritis. Rats of Group II (n = 23, nephrotoxic serum, NTS only) were preimmunized by subcutaneous injection of 5 mg of normal rabbit IgG in complete Freund’s adjuvant (5 mg/kg of body weight; Sigma Chemical Co., St. Louis, Mo., USA) 5 days prior to the induction of accelerated anti-GBM glomerulonephritis [30] (fig. 1). The rats were then injected intravenously with a bolus dose of 12.5 mg/kg of body weight of rabbit anti-rat GBM antibody (NTS) (Otsuka, Japan) to induce crescentic glomerulonephritis. Group III rats (n = 55, NTS + irradiation) received local bilateral kidneys X-ray irradiation using a Toshiba EXS-300 X-ray machine (200 kV, 15 mA apparatus with 0.5 mm Cu + 0.5 mm Al filter at a dose rate of 0.458 Gy/min), at 6, 13, 20, and 27 days after the bolus injection of NTS (NTS7dRa1d, NTS14dRa1d, NTS21dRa1d and NTS28dRa1d, respectively).

For kidney irradiation, both kidneys were exposed through a surgical incision and covered with sterile gauze saturated with physiological saline solution after intraperitoneal Nembutal anesthesia (25 mg/kg of body weight). A specially designed lead shield was used that protected the entire body but allowed irradiation of the exposed kidneys bilaterally. A single low dose of 0.5 Gy X-ray was applied. After irradiation, the kidneys were restored to their original position and the animals were provided with regular diet and water ad libitum.

Five or six rats were sacrificed on each time interval at 1 day post-irradiation, while rats of Group II were also euthanized at day 7, 14, 21, or 28 after NTS injection. Five rats of Group III were sacrificed by deep anesthesia at day 8, 15 or 22 post-irradiation.

In addition, rats of Group I were used as control animals at various time intervals. All the experimental procedures described in this study followed the guidelines for care and use of laboratory animals of Nagasaki University.

Renal Function Tests and Blood Leukocyte Count At the time of sacrifice, blood samples were collected from the superior vena cava of different groups of rats. The levels of blood urea nitrogen and serum creatinine were measured by an autoanalyzer (Hitachi 7170; Hitachi City, Japan). Blood leukocyte counts were determined using TOA Microcell Counter CC-108 (SLC, Inc., Japan).

Histological Examination Renal tissues were collected and then fixed immediately in either 10% neutral-buffered formalin for 24 h or Carnoy’s solution for 2 h, embedded in paraffin and cut into 4-␮m-thick sections and stained with periodic acid-Schiff (PAS), periodic acid-methenamine Sliver and Masson’s trichrome. Glomerular lesions and tubulointerstitial damage were scored as described previously [28]. Briefly, glomerular hypercellularity was evaluated as total glomerular cell counts/glomerular cross section (gcs) for 50 glomeruli per animal in PAS-stained sections, as follows: 0, normal (!50 cells/gcs); 1, mild (50–80 cells/gcs); 2, moderate (80–120 cells/gcs); 3, severe (1120 cells/gcs). The percentage of glomeruli with crescentic formation and glomerular global sclerosis were assessed by examining at least 100 glomeruli per animal on PASstained sections. Glomerular crescents were classified as cellular crescents, fibrocellular crescents and fibrous crescents. In addition, cortical tubulointerstitial injury was characterized by tubular dilation and atrophy, cast formation, inflammatory cell infil-

Radiotherapy and Progression of Crescentic Glomerulonephritis

Sacr*

Sacr

Sacr

Sacr

7

14

21

28

14

21

28

21

28

Ra*

Rabbit IgG

NTS*

Days –5

0

6

7

Preimmunizing

Ra 13 14 Ra 20 21

28 Ra 27 28

Fig. 1. Schematic diagram of the study protocol and the three

study groups. Ra = Radiation; Sacr = sacrifice.

trates into the interstitium and interstitial fibrosis. It was semiquantitative analyzed in Masson’s trichrome-stained sections and divided into four grades as follows: 0, no abnormal findings; 1+, mild (!30% of the cortex); 2+, moderate (30–60% of the cortex); 3+, severe (1 60% of the cortex).

Immunohistochemistry and TUNEL Assay Immunohistochemistry was performed as described earlier [31–36]. Briefly, paraffin sections (4 ␮m) were deparaffinized with xylene, rinsed thoroughly with ethanol, then soaked in 0.3% hydrogen peroxide (H2O2) in methanol for 30 min at room temperature (RT) in order to inactivate endogenous peroxidase activity. The sections were incubated with either 10% goat serum or 10% rabbit serum for 30 min, then covered with primary antibodies and incubated overnight at 4 ° C. The sections were washed with phosphate-buffered saline and processed further using Histofine SAB-PO kit (Nichirei, Tokyo), as directed by the manufacturer, and developed with 3,3ⴕ-diaminobenzindine and H2O2. Finally, the sections were counterstained with hematoxylin or methyl green and mounted in glycergel (Dako A/S, Glostrup, Denmark). The following antibodies were used in this study.

Monoclonal mouse anti-proliferating cell nuclear antigen (PCNA) (Dako A/S), ED1 (a specific monocyte/macrophage marker; Serotec, Oxford, UK), rabbit polyclonal anti-human/mouse caspase 3 active antibody (Techne, USA), active caspase 7 polyclonal antibody (MBL, Japan). Negative control sections were incubated with normal goat or rabbit serum at the same protein concentration as the primary antibody. The numbers of PCNA-positive cells and ED1-positive macrophages were determined in at least 50 gcs, as well as in 40 high-power fields of 1-mm2 cortical tubulointerstitial area selected at random, through the eyepiece of the light microscope. The immunohistochemical expression of active caspases 3 and 7 were examined under light microscopy. Sections were grouped into two patterns: negative or weakly stained sections were classified as negative, moderately and strongly stained sections as positive.

Am J Nephrol 2008;28:555–568

557

Table 1. Renal function and histological assessment of various experimental groups

Serum creatinine, mg/dl

Glomerular hypercellularity (0–3)

Glomerular crescents, %

Glomerular Tubulointerstitial global sclerosis, % damage (0–3+)

Group I

Control

0.2180.02

0.0380.02

0.080.0

0.080.0

0

Group II

NTS7d NTS14d NTS21d NTS28d

0.7280.02 1.7780.17 2.5580.15 2.6080.15

1.3380.21 1.9280.27 2.6780.21 2.3380.21

5.080.4 31.181.2 42.581.1 52.681.3

0.080.0 2.880.1 6.880.2 9.080.2

1+ 1+ 2+ 3+

Group III

NTS7dRa1d NTS14dRa8d NTS21dRa15d NTS28dRa22d NTS14dRa1d NTS21dRa8d NTS28dRa15d NTS21dRa1d NTS28dRa8d NTS28dRa1d

0.6980.05 1.3980.12* 1.0880.13** 0.9780.14** 1.7680.14 1.4880.11** 1.3580.04** 2.6480.09 2.6680.12 2.6680.06

1.1780.17 1.3380.21* 1.9280.08* 1.6780.21* 1.8380.17 2.0080.22* 1.7580.25* 2.5080.22 2.2580.17 2.2580.17

4.480.2 27.781.0* 37.783.5* 44.380.8** 30.981.2 40.580.8* 45.280.8* 42.080.9 51.981.0 52.481.4

0.080.0 3.180.3 7.180.2 9.280.3 2.880.1 7.380.2 9.380.1 7.280.2 9.480.1 9.180.2

1+ 1+ 1+ to 2+ 2+ to 3+ 1+ 2+ 2+ to 3+ 2+ 3+ 3+

NTS = Nephrotoxic serum; d = day; Ra = radiation. Data represent the mean 8 SEM for each group of 5–6 animals. * p < 0.05; ** p < 0.001, compared with rats of Group II of same time point, respectively.

The TUNEL method was used to detect in situ DNA fragmentation to substantiate the apoptosis identified by light microscopy.

For this experiment, the ApopTag쏐Plus peroxidase in situ apoptosis detection kit (S7101; InterGen Co., USA) was used according to the protocol provided by the manufacturer [20]. Briefly, paraffin-embedded 4-␮m sections fixed in 10% neutral-buffered formalin were deparaffinized and rehydrated in graded ethanol, and then digested by incubation with a protein-digesting enzyme (20 ␮g/ml) for 15 min at 37 ° C. Endogenous peroxidase was inactivated by 3% H2O2 in phosphate-buffered saline for 5 min at RT. The sections were then immersed in TdT reaction buffer, and incubated under a humidified chamber with TdT enzyme for 60 min at 37 ° C. The slides were transferred to the stop buffer at RT for 10 min to terminate the reaction. The sections were incubated with anti-digoxingenin peroxidase, conjugated for 30 min at RT, developed by using 3,3ⴕ-diaminobenzindine peroxidase substrate and counterstained with 0.5% methyl green. Positive and negative controls for TUNEL stain were obtained. The number of TUNELpositive cells was determined as mentioned above. All cells staining positive with the TUNEL assay were counted, which included brown-stained nuclei, lightly-stained nuclei and pyknotic nuclei with apoptotic bodies. Furthermore, TUNEL staining and ED1 immunostaining were performed on serial sections.

Western Blotting The renal cortical tissues were removed and frozen immediately. The tissues were then suspended in lysis buffer (10 mM phosphate buffer, 1 mM phenylmethylsulfonyl fluoride, 2 ␮g/ml leupeptin, 2 ␮g/ml chymostain, and 0.2% Triton X-100), and supernatants were collected and the protein concentration was measured with an ultraviolet/visible spectrophotometer (UV-1600;

558

Am J Nephrol 2008;28:555–568

Shimadzu, Japan). Equal amounts of protein samples (50–100 ␮g/ lane) were subjected to electrophoresis on 10 or 15% sodium dodecyl sulfate-polyacrylamide gel and transferred to polyvinylidene difluoride membrane (Amersham Life Science, Bucks., UK).

The membranes were blocked with 5% non-fat dry milk in TBST (50 mM Tris (pH 7.5) 150 mM NaCl, and 0.1% Tween 20) buffer at pH 7.5 for 1 h at RT. Immunoblot analyses were performed with the following primary antibodies: rabbit polyclonal anti-human/ mouse caspase 3 active antibody (Techne), active caspase 7 polyclonal antibody (MBL), and anti-p53 mouse monoclonal antibody, p53 (Al-1) (Oncogene Research Products, San Diego, Calif., USA). The membranes were incubated overnight at 4 ° C with primary antibodies, washed in TBST buffer and further incubated with sheep anti-mouse IgG or donkey anti-rabbit IgG coupled to horseradish peroxidase in TBST buffer for 1 h at RT by using RPN 2108 ECL Western blotting analysis system (Amersham Pharmacia Biotech Inc., Piscataway, N.J., USA). Immunoreactive protein was detected by enhanced chemiluminescence, using the protocol provided by the manufacturer. The ECL detected blots were exposed to Polaroid film using the ECL mini-camera (RPN 2069).

Expression of ␤-actin was used as a loading control.

Statistical Analyses Data were expressed as mean 8 SEM. Differences between groups were examined for statistical significance using the MannWhitney U test. A p value !0.05 indicated a statistically significant difference.

Liu /Nazneen /Taguchi /Razzaque

3.5

3.0

3.0 Glomerular hypercellularity

Serum creatinine (mg/dl)

2.5

2.0

*

1.5

**

*

1.0

** 0.5

**

*

2.0

*

1.5

*

1.0

*

*

0.5 0

0

–0.5

a

7

14

21

60

28 days

b

7

14

21

28 days

*

50 Glomerular crescents (%)

2.5

*

40

** Control

*

30

NTS only NTS7dRa

*

20

NTS14dRa

10 0 –10

c

7

14

21

28 days

Fig. 2. Serum creatinine (a), glomerular hypercellularity (b) and percent of glomerular crescents (c) in different

groups of rats. Note that NTS-treated rats have increased serum level of creatinine, associated with increased glomerular hypercellularity, and increased glomerular crescent formation. Radiation treatment can improve both structural and functional aspects of NTS-treated rats. Details of other parameters are summarized in table 1. * p ! 0.05; ** p ! 0.001.

Results

Renal Function and Blood Examination The level of serum creatinine in Group II began to increase from day 7, significantly increased from day 14 and reached maximum level at day 28 post-NTS injection, compared with the control (p ! 0.05, each). Serum creatinine levels of Group III at 1 day post-irradiation were not significantly different from those of Group II at the same time interval (p 1 0.05). Group III rats have received local kidney irradiation at 6, 13, 20, and 27 days after the bolus injection of NTS and will be referred as NTS7dRa1d, Radiotherapy and Progression of Crescentic Glomerulonephritis

NTS14dRa1d, NTS21dRa1d and NTS28dRa1d, respectively. For NTS7dRa1d and NTS14dRa1d rats of Group III, serum creatinine levels were significantly lower at days 8, 15, and 22 post-irradiation compared with Group II at the same time points (p ! 0.05 or 0.001). However, for NTS21dRa1d rats, no falls in serum creatinine levels were noted at day 8 post-irradiation compared to Group II at the same time interval (table 1; fig. 2). No apparent changes in leukocyte counts were noted in various experimental groups (data not shown).

Am J Nephrol 2008;28:555–568

559

a

b

c

d

e

f

Fig. 3. Histological changes in NTS-induced crescentic glomerulonephritis and effects of kidney irradiation after NTS injection. a Control sham-operated rat kidney with no abnormality. b, d and f: 14, 21, and 28 days after NTS injection in Group II, respectively. c, e and g: 1, 8, and 15 days post-irradiation treatment, irradiation was applied at day 13 after NTS injection in Group III, respectively. b Mild to moderate hypercellu-

larity of the glomerular tufts and focal fibrin deposition with early cellular crescent formation. c Mild hypercellularity of the glomerular tufts with adhesion to Bowmanâ&#x20AC;&#x2122;s capsule. d Moderate-to-severe hypercellularity of the glomerular tufts with marked cellular crescents (arrows), focal tubular degeneration and interstitial inflammation. e Mild mesangial proliferation with small fibrocellular crescent and tubular degeneration and interstitial inflammation. f Segmental glomerular proliferative/sclerotic lesions with fibrocellular crescent formation (arrows), tubular atrophy, extraglomerular inflammation and interstitial fibrosis. g Fibrous crescent formation and focal interstitial inflammation. PAS staining. Orig. magnif. !100.

Histological Changes Table 1 summarizes the histological findings in various experimental groups. No significant histological changes were noted in the control kidney throughout the entire experimental period (fig. 3a). Glomerular hypercellularity, crescents, global sclerosis and tubulointerstitial damage developed gradually throughout the experiment in Group II (table 1; fig. 3b, d, f). The percentage of crescents reached maximal level at day 28 post-NTS injection (NTS28d 52.6 8 1.3%). Changes in Group III at day 1 post-irradiation were not significantly different from those of Group II at the same time interval (p 1 0.05). For NTS7dRa1d and NTS14dRa1d rats of Group III, 560

Am J Nephrol 2008;28:555â&#x20AC;&#x201C;568

g

glomerular hypercellularity, crescent formation and glomerular global sclerosis were significantly lower at days 8, 15, and 22 post-irradiation than those of Group II at the same time intervals (table 1, p ! 0.05 or 0.001; fig. 3c, e, g). The extent of tubulointerstitial damage was also reduced. However, no significant difference in histological changes was noted at day 8 post-irradiation between NTS21dRa1d rats of Group III and those of Group II at the same time point.

Immunohistochemical Analyses and TUNEL Staining For immunohistochemical examination and TUNEL staining, we used renal tissues obtained from kidneys Liu /Nazneen /Taguchi /Razzaque

a

a

b

b

c

c

Fig. 4. PCNA immunostaining in rats with NTS-induced crescentic glomerulonephritis and post-NTS irradiated-kidneys. a Con-

trol rat kidney, showing a few PCNA-positive cells in the glomerular tufts and interstitium. b 14 days after NTS injection (NTS14d of Group II), showing many PCNA-positive cells in the glomerular tufts and tubulointerstitium. c At day 1 post-irradiation, in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III). Note the marked reduction of PCNA-positive cells in the glomerular tufts and interstitium compared with NTS14d of Group II. Orig. magnif. !132.

Radiotherapy and Progression of Crescentic Glomerulonephritis

Fig. 5. Immunostaining of ED1 macrophages in rats with NTS-

induced crescentic glomerulonephritis and post-NTS irradiatedkidneys. a Control rat kidney. Note the lack of ED1-positive cells. b 14 days after NTS injection (NTS14d of Group II). Note the presence of numerous ED1-positive macrophages in the glomerular tufts and tubulointerstitium. c At day 1 post-irradiation, in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III), note the marked reduction of the number of ED1-positive macrophages in glomerular tufts and tubulointerstitium compared with NTS14d of Group II. Orig. magnif. !132.

Am J Nephrol 2008;28:555â&#x20AC;&#x201C;568

561

of NTS7dRa1d, NTS14dRa1d, NTS21dRa1d and NTS28dRa1d Group III rats, because this time was closely related to radiation-induced apoptosis in this study. Sections of the corresponding groups of NTS injection only and control rats were also immunostained and examined for apoptosis by TUNEL staining. The number of PCNAand ED1-positive cells in the irradiated kidneys were markedly reduced compared with their corresponding Group II at the same time intervals (fig. 4, 5). The extent of immunostaining for active caspases 3 and 7 was increased in the irradiated kidneys compared with the corresponding Group II at the same time points (fig. 6, 7).

The number of TUNEL-positive cells in the irradiated kidneys was markedly higher than in Group II at the same time periods (fig. 8). Immunostaining for macrophages and TUNEL staining using serial sections of radiation-treated kidneys showed a few ED1-positive macrophages also stained by TUNEL (fig. 9). As shown in figure 10a–d, the numbers of PCNA- and ED1-positive cells were significantly lower in kidneys of NTS7dRa1d and NTS14dRa1d rats, compared with Group II at the same time points (p ! 0.05). No significant differences were noted in kidneys of NTS21dRa1d and NTS28dRa1d rats. As shown in figure 10e and f, the number of TUNEL-positive cells was significantly higher in kidneys of NTS7dRa1d and NTS14dRa1d rats, compared with Group II at the same time intervals (p ! 0.05). There were no significant differences in kidneys of NTS21dRa1d and NTS28dRa1d rats of Group III. Western Blotting for Active Caspase 3, Active Caspase 7 and p53 For Western blot analysis, renal cortical tissues were harvested from kidneys of NTS14dRa1d rats, the corresponding groups of NTS injection only and sham control rats. Western blot analysis showed a marked increase in the 17-kDa active caspase-3 band in NTS14dRa1d kidneys compared with that of rats of NTS injection only (fig. 11). Increased expression of the 35-kDa active caspase 7 was also detected in NTS14dRa1d kidneys (fig. 11).

A marked increase in the expression of p53 protein, which is closely related to radiation-induced apoptosis, was observed in NTS14dRa1d kidneys compared with the rat groups of NTS injection only at the same time intervals.

However, there were no significant differences between the kidneys of NTS injection only and sham-control rats (fig. 12).

562

Am J Nephrol 2008;28:555–568

Discussion

Our study evaluated the effect of irradiation of the kidney on the progression of established accelerated antiGBM glomerulonephritis in rats. In our model, glomerular injury, renal dysfunction and histological damage developed gradually throughout the experiment. The percentage of crescents reached maximal level at 28 days after NTS injection. Although the changes seen at day 1 post-irradiation were not significantly different from those seen in kidneys of rats injected with NTS alone, changes noted in the NTS7dRa1d and NTS14dRa1d kidneys (glomerular hypercellularity and crescent formation) decreased at 8, 15, or 22 days after radiation treatment (p ! 0.05 or 0.001). Changes in renal function paralleled the histological findings. As demonstrated by immunostaining for PCNA and ED1, the numbers of proliferating cells and macrophages in the irradiated kidneys at day 1 post-irradiation were markedly reduced compared with the corresponding rats of the NTS injection group, especially for the NTS7dRa1d and NTS14dRa1d groups (p ! 0.05). The results of TUNEL staining showed that the number of apoptotic cells in the irradiated kidneys at day 1 post-irradiation was markedly higher than in rats injected with NTS alone, and the rate of apoptosis was especially higher in NTS7dRa1d and NTS14dRa1d kidneys (p ! 0.05). The results of immunostaining for ED1 and TUNEL staining using serial sections showed a few apoptotic macrophages. Consequently, it appears that radiation therapy could prevent the early stages of development of rat crescentic glomerulonephritis by promoting apoptosis. Furthermore, an association between radiation-induced apoptosis in kidney cells (intrinsic and infiltrating) and improved outcome in an experimental anti-GBM nephritis model may be partly achieved by apoptosis of macrophages; such cell deletion may be an important mechanism for counterbalancing macrophage proliferation, in order to limit the extent of tissue damage during the inflammatory response. Of relevance, Yang et al. [37] demonstrated that caspase inhibition could reduce renal apoptosis and ameliorates inflammation to improve proteinuria in experimental glomerulonephritis.

The caspase family plays a central role in the execution of apoptosis. Among the caspases, caspase 3 has been implicated in radiation-induced apoptosis [38–40]. Caspase 3 is translated as an inactive 32-kDa precursor that is proteolytically processed to become a functionally active enzyme [41–44]. Activation of caspase 3 requires two proteolytic cleavage events: removal of the NH2 terminal Liu /Nazneen /Taguchi /Razzaque

a

a

b

b

c

c

Fig. 6. Immunostaining of active caspase 3 in rats with NTS-induced crescentic glomerulonephritis and post-NTS irradiatedkidneys. a Control rat. Note the negative staining for caspase 3. b 14 days after NTS injection (NTS14d of Group II); note the moderate staining. c At 1 day post-irradiation treatment, in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III), note the strong staining pattern. Orig. magnif. !100.

Radiotherapy and Progression of Crescentic Glomerulonephritis

Fig. 7. Immunostaining of active caspase 7 in rats with NTS-in-

duced crescentic glomerulonephritis and post-NTS irradiatedkidneys. a Control rats. Note the negative staining for caspase 7. b 14 days after NTS injection (NTS14d of Group II). Note the moderate staining pattern. c At day 1 post-irradiation, in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III).

Note the strong staining pattern. Orig. magnif. !100.

Am J Nephrol 2008;28:555â&#x20AC;&#x201C;568

563

8

a

b

c

d

Fig. 8. Detection of apoptosis on TUNEL

9

a

564

b

Am J Nephrol 2008;28:555â&#x20AC;&#x201C;568

in rats with NTS-induced crescentic glomerulonephritis and post-NTS irradiatedkidneys. a Control rat. Note the lack of apoptosis. b 14 days after NTS injection (NTS14d of Group II). Note the presence of a few apoptotic cells (arrows) in Bowmanâ&#x20AC;&#x2122;s epithelial cells and tubulointerstitium. c, d At 1 day post-irradiation treatment, in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III).

Note the increased apoptosis (arrows) in glomerular tufts and cellular crescent, especially in the tubulointerstitium, compared with NTS14d in Group II. Orig. magnif. !100.

Fig. 9. Immunostaining of macrophages ED1 and TUNEL staining on serial sections from rat kidney at 1 day post-irradiation in rats irradiated at day 13 after NTS injection (NTS14dRa1d of Group III). a ED1. b TUNEL. A few of macrophages ED1-positive cells were also stained by TUNEL. Orig. magnif. !80.

Liu /Nazneen /Taguchi /Razzaque

40 #

35

NS

NS 20 PCNA+ cells/mm2

30 PCNA+ cells/gcs

NS

25

NS

# 25 20 15 10

# 15 # 10 5

5

a

0

CNT

7

14

21 NS

18

28 days

b

0

CNT

7

NS

21

28 days

NS

NS

21

28 days

# 20

16

#

12

ED-1+ cells/mm2

14 ED-1+ cells/gcs

14

#

10 8 6

15 # 10

5

4 2

c

0

CNT

7

14

21

28 days

d

0

CNT

7

# 6 5

NS

9

NS

4 3 2

NS

8 TUNEL+ cells/mm2

TUNEL+ cells/gcs

#

10

#

14

NS

#

7 6 5 4 3 2

1

1

e

0

CNT

7

14

21

28 days

f

0

CNT

7

14

21

28 days

Fig. 10. Quantification of PCNA-, ED1-, and TUNEL-positive

e TUNEL-positive cells in glomeruli. f TUNEL-positive cells in

cells in rat kidneys at 1 day post-irradiation in rats irradiated at day 6, 13, 20, and 27 after NTS injection (Group III) and Group II on the same days. a PCNA-positive cells in glomeruli. b PCNApositive cells in the tubulointerstitium. c ED1-positive cells in glomeruli. d ED1-positive cells in the tubulointerstitium.

the tubulointerstitium. Open bars: control (Group I), grey bars: NTS only (Group II), black bars: radiation after NTS injection (Group III). Data are the mean 8 SEM for each group of 5â&#x20AC;&#x201C;6 animals. # p ! 0.05. NS = Not significant.

pro-domain, generating a 29-kDa processing intermediate, which is subsequently cleaved into 17- and 11-kDa or 12-kDa subfragments [41, 43, 45]. Caspase 3 participates in apoptosis by cutting off contacts between surrounding

cells, reorganizing the cytoskeleton, marking the cell for phagocytosis and disintegrating the cell into apoptotic bodies. In the present study, immunostaining for active caspases 3 and 7 was stronger in irradiated kidneys than

Radiotherapy and Progression of Crescentic Glomerulonephritis

Am J Nephrol 2008;28:555â&#x20AC;&#x201C;568

565

1

2

3

5

4

6

7

8

9

Caspase 3

Caspase 7

␤-Actin

Fig. 11. Western blot analysis identified the 17-kDa active caspase 3 protein, and marked increase in the expression in radiationtreated kidneys (Group III) compared with Group II at similar time intervals. Lanes 1–3: renal cortical tissues from kidneys at 1 day post-irradiation in rats irradiated at day 13 after NTS injection. Western blot analysis identified the 35-kDa active caspase 7 protein, and overexpression in radiation-treated kidneys (Group III) compared with Group II at similar time intervals. Lanes 4–6: renal cortical tissues from kidneys at 14 days after NTS injection.

Lanes 7–9: sham-control rat kidneys. The bottom panel is the loading control, showing ␤-actin expression.

1

2

3

4

5

6

7

8

9

10

p53

␤-Actin

Fig. 12. Western blot analysis identified p53 protein in radiationtreated kidneys, and marked increase in the expression in radiation-treated kidneys (Group III) compared with Group II at similar time intervals. The expression of p53 protein in Group II was similar to that in sham-control rats. Lanes 1–3: renal cortical tissues from kidneys at 1 day post-irradiation in rats irradiated at day 13 after NTS injection. Lanes 4–6: renal cortical tissues from kidneys at 14 days after NTS injection. Lanes 7–9: sham-control rat kidneys. Lane 10: positive control. The bottom panel is the loading control, showing ␤-actin expression.

in kidneys of rats of the NTS injection group. The results of Western blotting for active caspase 3 showed markedly increased expression, as well as overexpression of active caspase 7 in the radiation group compared with rats of the NTS injection only. These results suggest that 566

Am J Nephrol 2008;28:555–568

radiation-induced apoptosis in our experimental model is dependent on caspase 3 pathway. To our knowledge, this is one of the few studies that highlights the importance of caspase 7 in the radiation-induced apoptosis [46]. Compared with the sham control kidneys, increased expression of active caspases 3 and 7 was observed in rat crescentic glomerulonephritis. The importance of caspase 3 in the apoptosis machinery has been reported previously in a nephrotoxic nephritis model of experimental progressive glomerulonephritis [35].

The mechanism of radiation-induced apoptosis remains poorly understood. A large variety of stimuli (e.g., radiation, chemotherapeutic agents), both intracellular and extracellular, can initiate the cell death program. It is assumed that radiation kills cells secondary to mitotic cell death caused by DNA double-strand breaks [47, 48]. DNA damage acts as an important chemical stimulus to induce various cellular responses to ionizing radiation. Ionizing radiation results in DNA damage and a cascade of events that lead to repair of DNA damage and inhibition of transition into the S phase. Irradiation is also known to induce cell cycle arrest, allowing cells to repair some of the DNA damage induced by radiation. The tumor suppressor protein p53 plays a pivotal role in determining whether radiation-damaged cells undergo apoptosis. These results indicate that radiation-induced apoptosis might have an association with p53 expression. Phosphorylated p53 protein also binds the DNA helix at specific nucleotide sequences close to p53-regulated genes, modifying their transcription. p53 activation leads to cell cycle arrest, DNA repair or apoptosis. Although p53 plays a significant role in irradiation-induced apoptosis, p53-independent apoptosis is known to occur as well. Early apoptosis is reported to be p53-dependent, whereas late apoptosis is p53-independent [49, 50]. The results of Western blotting for p53 revealed no obvious differences between kidneys of rats of the sham-operated control and NTS injected only group. These results indicate that spontaneous apoptosis in experimental crescentic glomerulonephritis might not be a p53-regulated process. We speculate that the change in the ratio of Bax to Bcl-2 is involved in induction of apoptosis, including caspase activation, and modulation of renal apoptosis, associated with renal inflammation, tubular atrophy and renal fibrosis [51].

Radiotherapy is an important treatment for many cancers. The molecular response to radiotherapy is an important issue to explore. Exposure of cells to ionizing radiation results in immediate and widespread oxidative damage to lipid membranes, for example, and can induce apoptosis of specialized cell types; the most important Liu /Nazneen /Taguchi /Razzaque

subcellular target is DNA [52]. Further studies need to be performed to understand whether mechanism of radiation-induced apoptosis is specific and different than nephrotoxic drug-induced DNA damage [53, 54].

In conclusion, radiation treatment applied locally to the kidneys could inhibit the progression of rat crescentic glomerulonephritis by promoting the apoptotic process. Radiation-induced apoptosis has an association with expression of p53 in experimental crescentic nephritis. Further studies using p53 inhibitors will determine whether radiation-induced apoptosis in crescentic nephritis is a p53-de-

pendent process or not. Moreover, it appears likely that caspases 3 and 7 play pivotal roles in the process of radiation-induced apoptosis. The results may provide the basis for more effective anti-nephritis treatment protocols and highlight the need for more work in this area.

Acknowledgments The authors thank Ms. M. Ide and Ms. K. Egashira for the excellent technical assistance.

References 1 Razzaque MS, Lanske B: The emerging role of the fibroblast growth factor-23-klotho axis in renal regulation of phosphate homeostasis. J Endocrinol 2007;194:1–10.

2 Vidotti DB, Arnoni CP, Maquigussa E, Boim MA: Effect of long-term type 1 diabetes on renal sodium and water transporters in rats.

Am J Nephrol 2008; 28:107–114.

3 Razzaque MS: Can fibroblast growth factor 23 fine-tune therapies for diseases of abnormal mineral ion metabolism? Nat Clin Pract Endocrinol Metab 2007;3:788–789.

4 Miyamoto K, Ito M, Tatsumi S, Kuwahata M, Segawa H: New aspect of renal phosphate reabsorption: the type IIc sodium-dependent phosphate transporter. Am J Nephrol 2007; 27:503–515.

5 Razzaque MS: Does renal ageing affect survival? Ageing Res Rev 2007;6:211–222.

6 Ichinose K, Kawasaki E, Eguchi K: Recent advancement of understanding pathogenesis of type 1 diabetes and potential relevance to diabetic nephropathy. Am J Nephrol 2007; 27:554–564.

7 Razzaque MS, Azouz A, Shinagawa T, Taguchi T: Factors regulating the progression of hypertensive nephrosclerosis. Contrib Nephrol. Basel, Karger, 2003, vol 139, pp 173–186.

8 Razzaque MS, Taguchi T: Factors that influence and contribute to the regulation of fibrosis. Contrib Nephrol. Basel, Karger, 2003, vol 139, pp 1–11.

9 Tipping PG, Holdsworth SR: T cells in crescentic glomerulonephritis. J Am Soc Nephrol 2006;17:1253–1263.

10 Holzman LB, Wiggins RC: Consequences of glomerular injury. Glomerular crescent formation. Semin Nephrol 1991; 11:346–353.

11 Levy JB, Turner AN, Rees AJ, Pusey CD: Long-term outcome of anti-glomerular basement membrane antibody disease treated with plasma exchange and immunosuppression. Ann Intern Med 2001;134:1033–1042.

Radiotherapy and Progression of Crescentic Glomerulonephritis

12 Oguma S, Okazaki H: Effective radiation therapy for crescentic glomerulonephritis after renal transplantation. Transplant Proc 1990;22:2264–2268.

13 Chen LM, Godinez J, Thisted RA, et al: New scoring system identifies kidney outcome with radiation therapy in acute renal allograft rejection. Int J Radiat Oncol Biol Phys 2000;46:999–1003.

14 Kettritz R, Wilke S, von Vietinghoff S, Luft F, Schneider W: Apoptosis, proliferation and inflammatory infiltration in ANCA-positive glomerulonephritis. Clin Nephrol 2006; 65: 309–316.

15 Shimizu A, Masuda Y, Kitamura H, Ishizaki M, Sugisaki Y, Yamanaka N: Apoptosis in progressive crescentic glomerulonephritis.

Lab Invest 1996;74:941–951.

16 Roos WP, Kaina B: DNA damage-induced cell death by apoptosis. Trends Mol Med 2006;12:440–450.

17 Miller JB, Girgenrath M: The role of apoptosis in neuromuscular diseases and prospects for anti-apoptosis therapy. Trends Mol Med 2006;12:279–286.

18 Kerr JF, Wyllie AH, Currie AR: Apoptosis: a basic biological phenomenon with wideranging implications in tissue kinetics. Br J Cancer 1972; 26:239–257.

19 Lan HY, Mitsuhashi H, Ng YY, et al: Macrophage apoptosis in rat crescentic glomerulonephritis. Am J Pathol 1997;151:531–538.

20 Bristow RG, Benchimol S, Hill RP: The p53 gene as a modifier of intrinsic radiosensitivity: implications for radiotherapy. Radiother Oncol 1996;40:197–223.

21 Lu-Hesselmann J, van Beuningen D, Meineke V, Franke E: Influences of TP53 expression on cellular radiation response and its relevance to diagnostic biodosimetry for mission environmental monitoring. Radiat Prot Dosimetry 2006;122:237–243.

22 Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RW: Participation of p53 protein in the cellular response to DNA damage. Cancer Res 1991;51:6304–6311.

23 Takagi M, Absalon MJ, McLure KG, Kastan MB: Regulation of p53 translation and induction after DNA damage by ribosomal protein L26 and nucleolin. Cell 2005; 123: 49–63.

24 Kastan MB, Berkovich E: p53: a two-faced cancer gene. Nat Cell Biol 2007; 9:489–491.

25 Fernando P, Megeney LA: Is caspase-dependent apoptosis only cell differentiation taken to the extreme? FASEB J 2007;21:8–17.

26 Nicholson DW, Ali A, Thornberry NA, et al: Identification and inhibition of the ICE/ CED-3 protease necessary for mammalian apoptosis. Nature 1995; 376:37–43.

27 Steller H: Mechanisms and genes of cellular suicide. Science 1995;267:1445–1449.

28 Naito T, Razzaque MS, Nazneen A, et al: Renal expression of the Ets-1 proto-oncogene during progression of rat crescentic glomerulonephritis. J Am Soc Nephrol 2000; 11: 2243–2255.

29 Liu D, Razzaque MS, Nazneen A, Naito T, Taguchi T: Role of heat shock protein 47 on tubulointerstitium in experimental radiation nephropathy. Pathol Int 2002; 52:340–347.

30 Lan HY, Bacher M, Yang N, et al: The pathogenic role of macrophage migration inhibitory factor in immunologically induced kidney disease in the rat. J Exp Med 1997; 185: 1455–1465.

31 Razzaque MS, Foster CS, Ahmed AR: Role of connective tissue growth factor in the pathogenesis of conjunctival scarring in ocular cicatricial pemphigoid. Invest Ophthalmol Vis Sci 2003;44:1998–2003.

32 Razzaque MS, Foster CS, Ahmed AR: Role of collagen-binding heat shock protein 47 and transforming growth factor-␤1 in conjunctival scarring in ocular cicatricial pemphigoid. Invest Ophthalmol Vis Sci 2003; 44: 1616–1621.

33 Razzaque MS, Nazneen A, Taguchi T: Immunolocalization of collagen and collagenbinding heat shock protein 47 in fibrotic lung diseases. Mod Pathol 1998; 11: 1183– 1188.

Am J Nephrol 2008;28:555–568

567

34 Razzaque MS, Taguchi T: Collagen-binding heat shock protein 47 expression in anti-thymocyte serum-induced glomerulonephritis.

J Pathol 1997; 183:24–29.

35 Yang B, El Nahas AM, Thomas GL, et al: Caspase-3 and apoptosis in experimental chronic renal scarring. Kidney Int 2001; 60: 1765– 1776.

36 Zha Y, Le VT, Higami Y, Shimokawa I, Taguchi T, Razzaque MS: Life-long suppression of growth hormone-insulin-like growth factor I activity in genetically altered rats could prevent age-related renal damage. Endocrinology 2006; 147:5690–5698.

37 Yang B, Johnson TS, Haylor JL, et al: Effects of caspase inhibition on the progression of experimental glomerulonephritis. Kidney Int 2003;63:2050–2064.

38 Harvey KF, Harvey NL, Michael JM, et al: Caspase-mediated cleavage of the ubiquitinprotein ligase Nedd4 during apoptosis. J Biol Chem 1998;273:13524–13530.

39 Fuchs EJ, McKenna KA, Bedi A: p53-dependent DNA damage-induced apoptosis requires Fas/APO-1-independent activation of CPP32␤. Cancer Res 1997;57:2550–2554.

40 Afshar G, Jelluma N, Yang X, et al: Radiation-induced caspase-8 mediates p53-independent apoptosis in glioma cells. Cancer Res 2006;66:4223–4232.

568

41 Schlegel J, Peters I, Orrenius S, et al: CPP32/ apopain is a key interleukin-1␤ converting enzyme-like protease involved in Fas-mediated apoptosis. J Biol Chem 1996; 271: 1841– 1844.

42 Thornberry NA: Interleukin-1␤ converting enzyme. Methods Enzymol 1994; 244: 615– 631.

43 Fernandes-Alnemri T, Litwack G, Alnemri ES: CPP32, a novel human apoptotic protein with homology to Caenorhabditis elegans cell death protein Ced-3 and mammalian interleukin-1␤-converting enzyme. J Biol Chem 1994; 269:30761–30764.

44 Vassar R: Caspase-3 cleavage of GGA3 stabilizes BACE: implications for Alzheimer’s disease. Neuron 2007;54:671–673.

45 Sabbatini P, Han J, Chiou SK, Nicholson DW, White E: Interleukin-1␤ converting enzymelike proteases are essential for p53-mediated transcriptionally dependent apoptosis. Cell Growth Differ 1997;8:643–653.

46 Yang B, Harris KP, Jain S, Nicholson ML: Caspase-7, Fas and FasL in long-term renal ischaemia/reperfusion and immunosuppressive injuries in rats. Am J Nephrol 2007; 27:397–408.

47 Sokolov MV, Smilenov LB, Hall EJ, Panyutin IG, Bonner WM, Sedelnikova OA: Ionizing radiation induces DNA double-strand breaks in bystander primary human fibroblasts.

Oncogene 2005;24:7257–7265.

Am J Nephrol 2008;28:555–568

48 Hall EJ: Molecular biology in radiation therapy: the potential impact of recombinant technology on clinical practice. Int J Radiat Oncol Biol Phys 1994;30:1019–1028.

49 Kim IA, Bae SS, Fernandes A, et al: Selective inhibition of Ras, phosphoinositide 3 kinase, and Akt isoforms increases the radiosensitivity of human carcinoma cell lines. Cancer Res 2005;65: 7902–7910.

50 Jonathan EC, Bernhard EJ, McKenna WG: How does radiation kill cells? Curr Opin Chem Biol 1999; 3:77–83.

51 Yang B, Johnson TS, Thomas GL, et al: A shift in the Bax/Bcl-2 balance may activate caspase-3 and modulate apoptosis in experimental glomerulonephritis. Kidney Int 2002; 62:1301–1313.

52 Yarnold J: Molecular aspects of cellular responses to radiotherapy. Radiother Oncol 1997;44:1–7.

53 Razzaque MS: Cisplatin nephropathy: is cytotoxicity avoidable? Nephrol Dial Transplant 2007;22:2112–2116.

54 Taguchi T, Nazneen A, Abid MR, Razzaque MS: Cisplatin associated nephrotoxicity and pathological events. Contrib Nephrol. Basel, Karger, 2005, vol 148, pp 107–121.

Liu /Nazneen /Taguchi /Razzaque