DNA Fragmentation in Acute and Chronic Rejection After Renal Transplantation

DNA Fragmentation in Acute and Chronic Rejection After Renal Transplantation

DNA Fragmentation in Acute and Chronic Rejection After Renal Transplantation U. Ott, A. Aschoff, R. Fünfstück, G. Jirikowski, and G. Wolf ABSTRACT Acu...

815KB Sizes 2 Downloads 122 Views

DNA Fragmentation in Acute and Chronic Rejection After Renal Transplantation U. Ott, A. Aschoff, R. Fünfstück, G. Jirikowski, and G. Wolf ABSTRACT Acute and chronic rejections are important denominators for the long-term function of renal grafts. One important indicator of cell damage is enzymatic DNA fragmentation. To investigate possible mechanisms, the rate of DNA fragmentation (TUNEL staining), the expression of tissue transglutaminase II (a marker of advanced DNA damage), and 8-hydroxy2=-deoxyguanosine (8-OhdG), an indicator of oxidative injury of nucleic acids, were studied by immunohistochemistry. Semithin sections of renal biopsies revealed 23 patients to show acute interstitial rejections (Banff 97 IA, IB); eight patients, acute vascular rejection (Banff 97 IIA, IIB); and 20 patients, chronic allograft nephropathy (Banff 97 I to III). Correlations were calculated between apoptotic cells and serum creatinine at the time of biopsy and after 6 months. In acute rejection, the proximal tubular cells were apoptotic, particularly in regions with mononuclear infiltrates. In consecutive sections, these apoptotic tubular cells also showed damage by reactive oxygen species (positive 8-OhdG staining). Patients with acute interstitial rejection revealed the highest number of tubular DNA fragmentation (14.9 ⫾ 10.3) versus chronic allograft nephropathy (9.2 ⫾ 5.6) as TUNEL-positive cells per 80,000 ␮m2 (P ⬍ .05). Patients with acute vascular rejection showed a low degree of tubular apoptosis (6.8 ⫾ 5.1). There was no significant difference in glomerular DNA fragmentation between acute interstitial and chronic rejections: acute interstitial rejection ⫽ 7.1 ⫾ 5.9 versus chronic allograft nephropathy ⫽ 6.1 ⫾ 3.9 TUNEL-positive cells per 80,000 ␮m2. There was a significant negative correlation between the degree of tubular (P ⬍ .01) and glomerular (P ⬍ .05) apoptosis and the serum creatinine at the time of biopsy as well as after 6 months in all patients irrespective of the Banff class. However, there was heterogeneity in the correlation between renal function and the degree of apoptosis in the glomerular and tubular compartments in the various Banff classes. A positive correlation (P ⬍ .01) was observed between the degree of tubular apoptosis and serum creatinine at 6 months after biopsy among patients with acute vascular rejection (Banff 97 IIA, IIB). The present data revealed a high degree of tubular DNA fragmentation associated with oxidative stress in acute interstitial rejection. Nevertheless, apoptosis did not generally negatively influence future renal function and may be important to clear proliferating cells. Apoptosis may also play a different pathophysiological role depending on the type of rejection.

O

NE MECHANISM to regulate cell number, besides cell cycle control, is programmed cell death or apoptosis.1 Although apoptosis has been previously described in renal allograft rejection, in cyclosporine nephrotoxicity as well as in ischemia-reperfusion injury,2– 4 only a few studies have quantitatively assessed apoptosis correlating the findings with the Banff 97 classification of renal allograft pathology.5,6 In addition, many studies have relied solely on DNA fragmentation as the marker for apoptosis.7 Apoptosis, however, as defined by © 2007 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 39, 73–77 (2007)

morphological criteria, is not always accompanied by DNA fragmentation.7,8 Therefore the use of additional markers is necessary to characterize apoptosis. One of these markers is From the Departments of Internal Medicine III (U.O., R.F., G.W.) and Anatomy II (A.A., G.J.), Friedrich-Schiller University, Jena, Germany. Address reprint requests to Gunter Wolf, MD, Department of Internal Medicine III, Friedrich-Schiller University, Erlanger Allee 101, D-07740 Jena, Germany. E-mail: [email protected] 0041-1345/07/$–see front matter doi:10.1016/j.transproceed.2006.10.023 73

OTT, ASCHOFF, FU¨NFSTU¨CK ET AL

74

tissue transglutaminase II, which has been found to be expressed exclusively in apoptotic cells in the small intestine of mice.8 Furthermore, oxidative stress may be an initiator of apoptosis,9 but direct staining for markers of oxidative stress and apoptosis have not yet been studied in renal allograft pathology. The aim of this investigation was to immunohistologically characterize DNA fragmentation,10 and transglutaminase II8 and 8-hydroxy-2=-deoxyguanosine (8-OhdG)11 expression in renal biopsies from patients with acute interstitial and acute vascular rejection as well as samples from patients with chronic allograft nephropathy.

MATERIALS AND METHODS Patients Retrospectively, we studied biopsies of 23 patients with acute interstitial rejections (Banff 97 IA, IB), acute vascular rejection (eight patients, Banff 97 IIA, IIB), and 20 patients with chronic allograft nephropathy (CAN; Banff 97 I-III).5 Renal biopsy was performed by clinical indication with ultrasound observation, using the BIOPTYo instrument. Routine diagnosis was performed by an experienced nephropathologist (Prof Dr H.J. Gröne, University of Heidelberg, Germany). Use of biopsy material for further studies beyond routine diagnosis was approved by the local ethics committee. Serum creatinine was measured with the JAFFE rat method, using a commercially available kit on a multianalyzer (Beckman, Brea, Calif, USA). Follow-up serum creatinine values were available for 18 patients with acute interstitial rejections, 18 patients with CAN, and four patients with acute vascular rejection.

Histological Analysis Biopsy material immediately fixed in 5% phosphate-buffered paraformaldehyde was stored at 4°C. After fixation, biopsies dehydrated through ascending ethanol series were embedded in EPON 812. Serial semithin sections (0.5 ␮m thick) were cut on a Reichert Ultracut E microtome. Resin was removed by treatment of sections with sodium methoxide prior to rehydration and immunostaining as previously described.10 DNA fragmentation was detected by in situ end labeling with bromodeoxyuridine (BrdU) and terminal transferase, followed by BrdU immunohistochemistry. The reaction product was visualized with the peroxidase-antiperoxidase (PAP) method. Consecutive sections were immunostained with an antitransglutaminase II antibody (Upstate Biotechnology, Waltham, Mass, USA). Selected biopsies were also stained for 8-OhdG, a marker of oxidative stress (mouse monoclonal anti-8-OhdG antibody, Oxis, Portland, Ore, USA). Detection of antibodies was performed with the PAP method as previously described.8,10 For control incubation the primary antibody was omitted. Sections were dehydrated and coverslipped with Entellan (Merck, Darmstadt, Germany). Stained sections were evaluated with an Olympus BX50 microscope equipped with a digital camera DP10 and the IMOLYM software for semiquantitative analysis. Nuclei and cells were counted by an investigator unappraised of the origin of the sections in relation to a defined area (80,000 ␮m2). We only counted renal cells (glomerular or tubular cells) that were positive for DNA fragmentation and transglutaminase II on consecutive semithin sections. Staining for 8-OhdG was evaluated semiquantitatively (0, ⫹/⫺, ⫹, ⫹⫹, ⫹⫹⫹).

Statistical Analysis Data are reported as mean values ⫾ SEM. For statistical analysis the programs Excel 5.0 and SPSS were used. Multigroup comparisons were performed with the Kruskal-Wallis test, individual groups were analyzed with the Mann-Whitney test. Correlations were calculated with the Spearman rank test. A P value of ⬍.05 was considered significant.

RESULTS

In patients with acute interstitial rejection (Banff 97 IA, IB), many proximal tubular cells revealed DNA fragmentation; these cells also stained positive for transglutaminase II (Fig 1A and 1B). Staining for 8-OhdG was positive in nuclei with DNA fragmentation, suggesting oxidative stress (Fig 1C, Table 1). Not surprisingly, only a few glomerular cells stained positive for DNA fragmentation, transglutaminase II, and 8-OhdG in biopsies of patients with acute interstitial rejection (Fig 1D to 1F). Although there were many cells in addition to tubular cells in the interstitium that revealed DNA fragmentation, transglutaminase II positivity, and 8-OhdG staining, arterioles were negative (Fig 1G to 1I). The highest number of DNA fragmentation-stained nuclei was found in tubules of patients with acute interstitial rejection (Fig 2). Quantitative analysis showed no significant difference in glomeruli (Fig 2). Biopsies from patients with CAN showed positive TUNEL staining in tubular and glomerular cells (Fig 3A and 3D). The majority of positive glomerular cells were inside capillaries suggesting apoptosis of circulating blood cells and not of intrinsic glomerular cells. In serial sections, cells with DNA fragmentation also stained positively for transglutaminase and 8-OhdG (Fig 3B, 3C, 3E, and 3F). Correlations between the degree of apoptosis in tubular and glomerular compartments and the serum creatinine values at the time of biopsy were calculated using the Spearman rank test (Table 2). In addition, for selected patients serum creatinine measurements were available 6 months after biopsy (Table 2). Interestingly, pooled data from all patients irrespectively of the Banff class of rejection revealed a significant negative correlation between the degree of apoptosis and renal function at the time of biopsy and after 6 months (Table 1). There was no significant correlation between the degree of tubular apoptosis and renal function in patients with acute interstitial rejection. However, tubular apoptosis showed a positive correlation with renal function after 6 months in patients with acute vascular rejection (Table 2). DISCUSSION

Apoptosis is a type of cellular death that is biochemically different from necrosis. Whereas apoptosis of infiltrating inflammatory cells may serve to clear leukocytes, apoptosis of intrinsic renal cells may lead to loss of renal parenchymal cells. The number of acute allograft rejections is an important predictive factor in the long-term function of the transplanted kidney.12 In addition, CAN that includes some

DNA FRAGMENTATION AND REJECTION

75

Fig 1. (A–I) Biopsy of a patient with acute interstitial rejection (Banff 97 IB). Staining of serial sections for DNA fragmentation (A, D, G), transglutaminase II (B, E, H), and 8-OhdG (C, F, I). In general, TUNEL-positive cells also stained for transglutaminase II and 8-OhdG. The majority of positive cells were tubules and the interstitium, whereas only a weak staining with few apoptotic cells was present in glomeruli. Vasculature itself showed a negative staining, but interstitial cells were positive (G, H, I). Magnification ⫻800. Bar ⫽ 20 ␮mol/L.

immunological aspects of chronic rejection is the single most common cause of late allograft loss. Although apoptosis of intrinsic kidney cells has been previously described in renal allografts during acute rejection,13,14 CAN, ischemia and reperfusion injury,2,4 and drug toxicity,3,15 only a few studies have assessed apoptosis quantitatively examining various anatomical compartments. Moreover, many studies have relied exclusively on TUNEL staining for detection of apoptosis.7 However, as neither DNA fragmentation nor internucleosomal DNA fragmentation are specific for apoptosis, we counted only nuclei with DNA fragmentation that exhibited a positive staining for transglutaminase II on serial sections. Activation of transglutaminase II is a later consequence of apoptosis, leading to the formation of intracellular protein cross-links, probably involved in the formation of apoptotic envelopes or apoptotic bodies to ensure that once apoptosis has been initiated, it will be finished without causing inflammation and apparent tissue injury.16 There is also a paucity of studies correlating apoptosis with renal function. Table 1. Qualitative Intensity of 8-OhdG Staining Tubules Interstitium Glomeruli Vasculature

Acute interstitial rejection CAN Acute vascular rejection

⫹⫹⫹ ⫹⫹ ⫹

⫹⫹ ⫹ ⫹⫹

⫹ ⫹⫹ ⫺

⫹ ⫺ ⫹⫹⫹

⫺, no staining; ⫹, slightly positive staining; ⫹⫹, strong positive staining; ⫹⫹⫹, very strong positive staining.

We found a high degree of tubular TUNEL-positive cells that also stained for transglutaminase II and 8-OhdG in patients with acute rejection. These findings link apoptosis

Fig 2. Quantification of DNA fragmentation in various forms of allograft pathology. Only cells that also stained positive for transglutaminase II in serial sections were counted in a defined area of 80,000 ␮m2 by an investigator unapprised of details. There were significantly more nuclei with DNA fragmentation tubules in patients with acute interstitial rejection compared with acute vascular rejection and CAN. There was no significant difference in glomerular nuclei with DNA fragmentation among the three groups. *P ⬍ .05 versus CAN, and acute vascular rejection; n ⫽ 23 acute interstitial rejection; n ⫽ 8 acute vascular rejection; and n ⫽ 20 CAN.

OTT, ASCHOFF, FU¨NFSTU¨CK ET AL

76

Fig 3. (A–I) Example of a patient with CAN (Banff 97 II). Staining of serial sections for DNA fragmentation (A, D, G), transglutaminase II (B, E, H), and 8-OhdG (C, F, I). Compared with acute tubular rejection, there was less tubular apoptosis, although clearly present. There was an increase in TUNEL-positive cells localized to glomeruli (D). However, the majority of these cells is inside the glomerular capillary and likely represent circulating blood cells and not intrinsic glomerular cells (D, E, F). Magnification ⫻800. Bar ⫽ 20 ␮mol/L.

with oxidative stress. Apoptotic tubular cells were found adjacent to regions of mononuclear infiltrates (data not shown). It has been previously shown that tubular cells express fas and fas-ligand.14,17 Tubular fas expression is, for example, induced by ␥-interferon.17 Therefore, it is possible that induced expression of fas antigen may initiate tubular apoptosis. However, some studies have shown that tubular fas expression increases nonspecifically posttransplantation and that fas-ligand expression is actually down-regulated suggesting that this system may not play a role in apoptosis of tubular cells during acute rejection.14 Moreover, it has been shown that tubular epithelium resists anti-fas antibody-mediated apoptosis, and that infiltrating lymphocytes primarily utilize perforin-dependent mechanisms of tubular apoptosis during acute allograft rejection.18 Not surprisingly, there were only few glomerular cells with signs

of DNA fragmentation in acute rejection because this process mainly involves tubules with tubulitis and apoptosis. Tubular apoptosis showed no significant correlation with renal function in patients with acute interstitial rejection. However, in patients with acute vascular rejection, the initial degree of tubular apoptosis was predictive for a decrease in renal function 6 months after biopsy. In contrast to acute rejection there were less tubular, but more glomerular cells with signs of apoptosis in patients with CAN. Although cyclosporine contributes to mononuclear cell apoptosis,19 the situation is more complex for tubular cells. Earlier studies indicate that cyclosporine in vivo stimulates tubular apoptosis through the up-regulation of fas-ligand.15 In contrast, a cell culture approach has provided evidence that in vitro cyclosporine does not directly mediate apoptosis, but at higher concentrations the

Table 2. Correlation Between Apoptosis and Serum Creatinine at the Time of Biopsy and After 6 Months TUNEL-positive Cells/80,000 ␮m2 All

Acute Interstitial Rejection

Glomerular

Serum creatinine at time of biopsy Serum creatinine 6 months after biopsy

Tubuli

Glomerular

Tubuli

P

R

P

R

P

R

P

R

.05

⫺.38

.01

⫺.50

.01

⫺.72

.19

⫺.32

.05

⫺.39

.01

⫺.71

.05

⫺.49

.10

⫺.39

Chronic Allograft Nephropathy

Acute Vascular Rejection Glomerular P

1.0 .90

Tubuli

R

P

.10

.20

.40

.01

Correlations were calculated with the Spearman rank test. R ⫽ correlation coefficient; P ⫽ two-sided significance.

R

.80 1.0

Glomerular

Tubuli

P

R

P

R

.74

.087

.06

⫺.47

.53

.15

.01

⫺.78

DNA FRAGMENTATION AND REJECTION

vehicle cremophore could induce this effect.20 These studies do not necessarily contradict and cyclosporine A has been shown to stimulate oxidative stress in various cells and could mediate apoptosis through reactive oxygen species.21 In this regard, tubular cells in CAN demonstrated a strong staining for 8-OhdG indicating ongoing oxidative stress. Interestingly, Pardo-Mindán and colleagues reported that mycophenolate mofetil (MMF) exhibits antiapototic properties.22 This may, at least to some extent, explain the reduced incidence of CAN under MMF therapy compared with calcineurin inhibitors. The increase in glomerular cells with positive TUNEL and transglutaminase II staining in patients with CAN was mainly due to apoptosis in circulating cells in capillaries, and only to a lesser extent due to intrinsic glomerular cells. This is also reflected by a negative correlation between apoptosis and serum creatinine after 6 months, suggesting that effective apoptotic clearance of proliferating, presumably inflammatory cells, is associated with better renal function. Our correlation analysis between the degree of apoptosis and renal function revealed heterogeneous data. Depending on the type of rejection and the anatomic department studied, the degree of apoptosis was either positively or negatively correlated with renal function. These findings suggested that apoptosis per se is not detrimental for renal function. In contrast, it may even serve to facilitate remodeling by clearance of damaged cells and by limiting proliferation of infiltrating inflammatory cells. Thus, apoptosis must be always considered in the context of the Banff class of rejection and the exact anatomic compartment. A limitation of our study is its retrospective nature and the relatively small number of patients. Unfortunately, some patients were also lost during follow-up. Nevertheless, we think our investigation is interesting because we meticulously quantified the degree of apoptosis with different markers, looked at a marker of oxidative stress, and correlated these findings with renal function.

REFERENCES 1. Savill J: Regulation of glomerular cell number by apoptosis. Kidney Int 56:1216, 1999 2. Laine J, Etelämäki P, Holmberg C, et al: Apoptotic cell death in human chronic renal allograft rejection. Transplantation 63:101, 1997 3. Rawat S, Gray C, Johnson TS, et al: Apoptosis and expression of BCL-2 and BAX in cyclosporine-induced experimental renal fibrosis. Transplant Proc 35:187, 2003

77 4. Daemen MA, de Vries B, Buurman WA: Apoptosis and inflammation in renal reperfusion injury. Transplantation 73:1693, 2002 5. Racusen LC, Solez K, Colvin RB, et al: The Banff 97 working classification of renal allograft pathology. Kindey Int 55:713, 1999 6. Kato S, Akasaka Y, Kawamura S: Fas antigen expression and its relationship with apoptosis in transplanted kidney. Pathol Int 47:230, 1997 7. Noronha IL, Oliveira SG, Tavares TS, et al: Apoptosis in kidney and pancreas allograft biopsies. Transplantation 79:1231, 2005 8. Aschoff AP, Günther G, Jirikowski GF: Tissue transglutaminase in the small intestine of the mouse as a marker for apoptotic cells. Colocalization with DNA fragmentation. Histochem Cell Biol 113:313, 2000 9. Redondo-Horcajo M, Lamas S: Oxidative and nitrosative stress in kidney disease: a case for cyclosporine A. J Nephrol 18:453, 2005 10. Aschoff A, Jantz M, Jirikowski GF: In-situ end labelling with bromodeoxyuridine—advanced technique for the visualization of apoptotic cells in histological specimens. Horm Metab Res 28:311, 1996 11. Pflaum M, Will O, Epe B: Determination of steady-stae levels of oxidative DNA base modifications in mammalian cells by means of repair endonucleases. Carcinogenesis 18:2225, 1997 12. Oberbauer R, Rohmoser M, Regele H, et al: Apoptosis of tubular epithelial cells in donor kidney biopsies predicts early renal allograft function. J Am Soc Nephrol 10:2006, 1999 13. Wever PC, Aten J, Rentenaar RJ, et al: Apoptotic tubular cell death during acute allograft rejection. Clin Nephrol 49:28, 1998 14. Porter C, Ronan JE, Cassidy MJ: Fas-fas-ligand antigen expression and its relationship to increased apoptosis in acute renal transplant rejection. Transplantation 69:1091, 2000 15. Shihab FS, Andoh TF, Tanner AM, et al: Expression of apoptosis regulatory genes in chronic cyclosporine nephrotoxicity favors apoptosis. Kidney Int 56:2147, 1999 16. Fésüs L, Szondy Z: Transglutaminase 2 in the balance of cell death and survival. FEBS Lett 579:3297, 2005 17. Du C, Jiang J, Guan O, et al: Renal tubular epithelial cell self-injury through Fas/Fas ligand interaction promotes renal allograft injury. Am J Transplant 4:1583, 2004 18. Wever PC, Boonstra JG, Laterveer JC, et al: Mechanism of lymphocyte-mediated cytotoxicity in acute renal allograft rejection. Transplantation 66:259, 1998 19. Andrikos E, Yavuz A, Bordoni V, et al: Effect of cyclosporine, mycophenolate mofetil, and their combination with steroids on apoptosis in a human cultured monocytic U937 cell line. Transplant Proc 37:3226, 2005 20. Bakker RC, van Kooten C, van de Lagemaat-Paape ME, et al: Renal tubular epithelial cell death and cyclosporin A. Nephrol Dial Transplant 17:1181, 2002 21. Chen HW, Chien CT, Yu SL, et al: Cyclosporine A regulate oxidative stress-induced apoptosis in cardiomyocytes: mechanisms via ROS generation, iNOS and Hsp70. Br J Pharmacol 137:771, 2002 22. Pardo-Mindán FJ, Errasti P, Panizo A, et al: Decrease of apoptosis rate in patients with renal transplantation treated with mycophenolate mofetil. Nephron 82:232, 1999