Partial nephrectomy, the standard of care for renal tumors under 4 cm [1], is a surgical operation of many effects, mainly because it applies two significant stress conditions in the kidney: ablation and ischemia–reperfusion. When the procedure is performed in a solitary kidney, the remaining renal function must not be impaired. Unfortunately, the only parameter that the surgeon can change to prevent such an event is the ischemia time, by clamping the renal vessels. In difficult cases, the applied ischemia can be prolonged considerably, causing detrimental damage to a patient with a solitary kidney, and condemning him/her to renal dialysis.Several clinical and laboratory studies have been per-formed in an effort to minimize the ischemia–reperfu-sion injury, either by trying novel surgical approaches or by administering protective agents. In the largest published study in humans, the proposed tolerable warm ischemia (WI) time limit reached the 20 min, while the cold ischemia (CI) limit was 35 min [2].By creating an experimental animal model that simu-lates the open partial nephrectomy in the solitary
kidney, we tried to challenge these limits and compare the two types of ischemia by recording the preserva-tion of renal function, which is directly related to the preservation of cellular structures. Particularly crucial for the renal function is the filtration barrier, which consists of 3 distinct layers: the podocytes, the glomer-ular basement membrane (GBM), and the endothelial cells. Since the cellular basis of renal failure is based in the injury of the above layers [3,4], we tried to investi-gate these effects of prolonged WI and CI using light and transmission electron microscopy (TEM).
Materials and Methods
experimental Protocol
Selection and preparation of the animal model, the anesthesiological and surgical protocol of the opera-tions, the euthanization protocol, the biochemical determinations, and the tissue preparation for the optical microscopy have been described in a previous
Received 04 October 2010; accepted 22 November 2010
Correspondence: Stavros Tyritzis, MD, PhD, Department of Urology, Athens University Medical School, LAIKO Hospital, 17 Agiou Thoma str., 11527, Athens, Greece. E-mail: [email protected]
Original article
effects of Prolonged Warm and Cold ischemia in a solitary Kidney animal Model after Partial nephrectomy:
an Ultrastructural investigationStavros I. Tyritzis, MD, PhD1, Michael Zachariades, PhD2, Kostas Evangelou, MD, PhD3,
Vassilis G. Gorgoulis, MD, PhD3, Aspasia Kyroudi-Voulgari, MD, PhD3, Kitty Pavlakis, MD, PhD4, Theodoros G. Troupis, MD, PhD2, and Constantinos A. Constantinides, MD, PhD1
1Department of Urology, Athens University Medical School-LAIKO Hospital, Athens, Greece, 2Department of Anatomy, Athens University Medical School, Athens, Greece, 3Department of Embryology–Histology, Athens University Medical
School, Athens, Greece, and 4Department of Pathology, Athens University Medical School, Athens, Greece
aBstraCt
Ischemia–reperfusion injury can be detrimental to the solitary kidney, especially when it is accompanied by ablation. To the authors’ knowledge, the effects of partial nephrectomy with prolonged application of ischemia have never been described at the ultrastructural level. Therefore, the authors used an animal model and focused on putative structural effects in the glomerular basement membrane and the podocytes. They demonstrate the advantageous role of cold ischemia, even in up to 120 min. In contrast, more than 60 min of warm ischemia leads to catastrophic lesions in all the cellular structures, as is reflected by mortality due to acute renal failure.
Keywords: Electron microscopy, Ischemia–reperfusion injury, Partial nephrectomy, Rabbit, Solitary kidney
paper by the authors [5]. The experimental flowchart is depicted in Figure 1. The protocol was approved by the Institutional Research and Ethics committee of the ‘Laikon’ Hospital of Athens (Ref. No. 464), the Biomedical Research Foundation of the Academy of Athens (Ref. No. A.03.1/4/10–05), and the competent Veterinary Service of the prefecture of Athens (Ref. No. K/5447).
light Microscopy
The hematoxylin–eosin-stained slides were evaluated by two pathologists blindly and findings were strati-fied by intensity and extent, using a customized scor-ing system by the authors. Interobserver variation was resolved by simultaneous reevaluation. Scoring ranged from 0 to 3 for both categories. The observations of no lesions, minor edema, major edema, atrophy, pyknotic nuclei, cast formation, and necrosis scored 0, 1, 2, and 3 points of intensity, respectively. Findings observed in less than 5, 5–25, 25–50%, and more than 50% of each section were given 0, 1, 2, 3 points of extent, respec-tively. Overall score (range 0–6) resulted from the sum-mation of the two individual scores. Overall scoring was defined as 0 (sum 0 to 1), 1 (sum 2 to 3), 2 (sum 4), and 3 (sum 5 to 6).
teM
Transmission electron microscopy was performed as previously described [6]. Briefly, the tissue was fixed in 4% glutaraldehyde (25% EM), cut into small blocks, and postfixed in 1% osmium tetroxide (SERVA). The tissue blocks were dehydrated in ethanols and in epoxypropane (Ferak) and infiltrated with araldite
(SERVA) that was finally polymerized. Semithin (1 μm thick) and thin (light gold interference color, about 90 nm thick) sections were cut from each block, using an ultramicrotome (Leica Ultracut). Semithin sec-tions were stained with toluidine blue and examined under a light microscope. Thin sections were collected on copper Formvar-coated grids, contrasted with ura-nyl acetate (Merck) and lead citrate (BDH), and exam-ined with a transmission electron microscope (JEOL 200CX) at 120 kV.
statistics
Intensity and extent scores were compared using the Fisher’s exact test. When necessary, Bonferroni adjust-ment for multiple comparisons was used and corrected p values are presented. Level of significance was set at a = .05. All statistical analyses were conducted using the SPPS 14. No statistical analysis was performed for the TEM observations due to lack of distinct quantification of the recorded lesions.
resUlts
light Microscopy
The intensity and extent scores for all groups are summarized in Table 1. Major abnormalities were detected in the 90-min WI sections (Figure 2). In contrast, a few mild lesions were observed in groups B, C, and D (Figure 2). Intensity of tissue damage was significantly different between the groups. The extent of the tissue damage exceeded 5% in all cases. Distribution of the extent of damage did not differ significantly between the groups. When the overall
score of intensity and extent of the damage was con-sidered, differences between groups became more pronounced.
teM
Sham GroupAt the ulrastructural level the normal glomerulus (Figure 3) reveals the following well-known features. Τhe podocytes have irregular shape with long exten-sions surrounding the arterioles of the glomerulus, while their foot processes are formatting one layer of the GBM. The nucleus of the podocyte is lobu-lar or asymmetrical and in some cells there is also a small nucleolus. The cytoplasm is dense, having
a plethora of organelles. Multiple mitochondria, Golgi apparatus, and well-organized endoplasmic reticulum are distributed inside the cytoplasm. The foot processes are a distinct entity, including cyto-plasm with organelles that is covered by a separate membrane. The interdigitation of the foot processes is linked via wide porous slits commonly known as slit diaphragms. The slit diaphragms are the exten-sions of the membrane that cover each foot process, as they do not form a characteristic double-lipid layer membrane but a fibrotic tape of protein. The GBM demonstrates its typical trilamellar structure (Figure 3). The foot processes are vertically placed externally, while internally, there is a fine layer of cytoplasm originating from the endothelial cell that forms the arteriole. The endothelial cells are irregular with a lobular nucleus.
Wi Groups
In general, prolonged WI induced severe cellular lesions, affecting primarily the GBM. The sever-ity of the lesions was dependent on the duration of ischemia.
FIGURE 2 Paraffin-embedded sections under light micros-copy. (a) 90-min WI: major abnormalities consisting of tubular dilation, atrophy and necrosis of the epithelial cells, fibrin cast and bloodcast formation, intraluminal aggregation of cells, due to exfoliation in the tubular lumen (arrows); (b) 60-min WI: extensive inflammation with viable tubules and glomeruli; (c) 90-min CI: mild lesions, mainly concerning the proximal and distal tubules, such as denuded basement membranes (arrows) and some pyknotic nuclei; (d) 120-min CI: excellent preservation of renal structures despite the prolonged applica-tion of ischemia. Magnifications: (a–c) ×20; (d) ×40.
FIGURE 3 Sham group under TEM. (a) Panoramic image of a normal glomerulus (magnification ×4700) including GBM (arrows), red blood cells (RBC), and podocytical nuclei (aster-isk); (b) typical presentation of the normal cytoplasm with endoplasmic reticulum, Golgi apparatus, and microtubules (magnification ×35,000).
TABLE 1 Intensity and extent scores of the recorded lesions in all study groups under light microscopy.No. (%) 90 min WI 60 min WI 90 min CI 120 min CI p valueIntensity Mild edema 0 (0) 5 (62.5) 3 (37.5) 6 (65) Severe edema, pyknotic nuclei, cast formation
Sixty minutes of WI did not cause irreversible lesions (Figure 4). The podocytes preserved their integrity, but lost their inner membrane and their mitochondria had poor inner structure. The GBM was well preserved, despite the fact that a distinct trilamellar structure was not evident in its entire extent and there was a partial loss of cellular polarity. Disintegration of the
slit diaphragms at an initial phase was also demon-strated. The endothelial cells were exhibiting primary signs of cellular necrosis. Additionally, we observed distortion and reduction in the quantity of intracellular organelles.
On the other hand, 90 min of WI cause detrimental effects at the ultrastructural level (Figure 4). Total loss of cellular differentiation with complete atrophy of the board-in-brush, necrosis, apoptotic bodies, extensive cellular edema, deterioration of the endothelial cells, flattening of the GBM, and foot process effacement are the main observations, which reflect the 100% mortality of the animals due to acute renal failure. The cytoplasm exhibited degeneration, and an increased number of autophagolysosomes was produced and expelled into the lumen of the arterioles. Dense nuclei with chromatin margination were also prominent.
It is noticeable that the appearance of the glomeruli was closely related to the postoperative course of the animals and their relevant renal function.
Ci Groups
No significant lesions were observed in the CI groups (Figure 5). Only a few podocytes were affected, and there was swelling and elongation of the mitochondria and the endoplasmic reticulum’s membrane, which was dilated. The normal trilamellar structure of the GBM is nicely preserved, although some of the slit diaphragms were not visible. This result was in accordance with the overall 100% survival of the animals, which had a normal renal function postoperatively, as determined by the biochemical determinations.
FIGURE 4 90 minutes WI: (a) deterioration of the GBM (magnification ×35,000); (b, c) detrimental lesions, no typical GBM, necrosis of the podocytes, and cytoplasmic dete-rioration (magnification ×4700). 60 min WI (d) very good preservation of the trilamellar GBM structure (magnification ×25,000); (e, f) podocyte with its nucleus in excellent shape (magnification ×4700).
FIGURE 5 90 min CI: (a) GBM in very good condition (mag-nification ×48,000); (b) excellent image of lampbrush tips in the proximal convoluted glomerulus (magnification ×72,000). 120 min CI: (c) near normal appearance of the GBM (magnifica-tion ×25,000); (d) excellent preservation of the cytoplasm with endoplasmic reticulum, Golgi apparatus, and microtubules (magnification ×35,000).
Our previous reports have attempted to establish the maximum tolerable time limit of WI and CI in this soli-tary kidney experimental model of animals undergoing partial nephrectomy and to compare these two types of ischemia in terms of their effect [5]. The main conclusion was that 60 min of WI and as long as 120 min of CI could be tolerated by the solitary rabbit kidney even when ablated. Moreover, the basic differentiation between the study groups was that the renal failure along with the catastrophic events happened only in the 90-min WI group. Conversely, the other three groups were character-ized by the same pattern of well-preserved normal renal function and renal structures in the light microscope, having only reversible lesions. Of course, hypothermia was advantageous independently of the applied time of ischemia.
To our knowledge, this is the first study demonstrat-ing the ultrastructural effects of CI during partial neph-rectomy, since the majority of published reports use cadaveric donors or the ischemic, ablated model without any reperfusion. Our model examined the late effects of ischemia–reperfusion injury, 15 days after reperfusion, and not the early effects occurring during the first 48 h of acute renal injury. This was performed deliberately because the majority of the ischemia–reperfusion models investigate the acute phase. The late phase could be more crucial in determining the possibility of recovery from an ischemic injury.
Every lesion affecting the glomerular filtration bar-rier is directly associated with the impairment of the renal function. However, it is not quite clear which lesion could have more impact or be more irreversible after such long periods of ischemia.
Based on the above data, we focused our attention on the alterations of the GBM and the podocytes, since it is well established that these are the key structures regu-lating glomerular filtration [4]. Moreover, the podocytes are highly differentiated cells with minimum capacity for regeneration [7] that not only are capable of synthe-sizing numerous chemokines and growth factors [7] but also form the final filtration barrier by structuring the slit diaphragms. We should again stress the fact that this study was a descriptive one, trying to clarify the major effects on the prementioned structures caused by prolonged WI, CI, and reperfusion.
Previous studies have described the various structural changes in the podocytes in challenging situations [8–20]. These changes include mostly foot process effacement, cell body attenuation, pseudocyst formation, and fusion of the podocytes to the GBM. The cytoplasm and nucleus of the podocyte are also disorganized and extended necrosis of the organelles is seen. The mitochondria exhibit severe edema and elongation. Formation of reactive oxygen species (ROS) and enhanced lipid peroxidation could be some of the major causes of podocyte effacement after an ischemia–reperfusion insult [21].
Several other alterations have been reported concern-ing the GBM, which loses its polarity due to proteo-glycan destruction and reduces its size selectivity [12]. Major lesions also include necrosis of the endothelium and the surrounding pericytes causing degeneration of the peritubular capillary endothelium and increased permeability [13]. Massive neutrophil infiltration is another common observation, which is regulated by numerous inflammatory mediators and stimulates the production of reactive oxygen species [15].
Our results were similar to the above. These, however, applied only to the 90-min WI group, since in the other three groups all the structures were very well preserved. Electron microscopy did not reveal any major lesions in podocyte or GBM morphology after 60 min of WI. This might indicate that the loss of cellular polarity and selec-tivity might be due to alterations in less-visible structures, such as proteoglycans and glycosaminoglycans [12]. It is also noticeable that in both CI groups, the status was near normal, when compared to the sham-operated group.
Possible theories explaining these detrimental events in the ultrastructural level can be drawn by the recorded results: (1) obstruction of the tubular lumen by debris, (2) obstruction of the peritubular capillaries by trapped erythrocytes, (3) no-reflow phenomenon due to severe endothelial damage [17], (4) destruction of the slit-diaphragms and loss of the typical trilamel-lar structure of the filtration barrier, and (5) obstruc-tion of the GBM due to the fusion with the podocytes. Another suggestion might have been the extreme vasoconstriction during the reperfusion state, causing shutdown of the filtration [18,19]. In general, all of the above co-existed, even though their quantification was not a subject of this study. Nonetheless, this issue could stimulate extremely interesting research.
We should acknowledge some issues that arose in this study. It is obvious that our results should not be applied to humans, since there are major anatomical differences between the human and the rabbit. Given that the ultrastructural findings were so vast and might not have been accurately quantified, we did not proceed to a quantitative analysis, despite the fact that some authors have reported scoring systems at the light and electron microscope level [22].
ConClUsion
Ischemia and reperfusion injury affect greatly the GBM, the podocytes, the endothelial cell layer, and the mese-chymatic structures. Yet, our results indicate that GBM and podocyte lesions have greater impact on the renal function. Possible protective agents of these two highly differentiated structures might greatly assist in cases of difficult renal tumors in which longer ischemia times are needed. Additionally, we also established that CI offers optimal glomerular preservation during a partial nephrectomy.
Ultr
astr
uct P
atho
l Dow
nloa
ded
from
info
rmah
ealth
care
.com
by
178.
128.
119.
247
on 0
2/07
/11
For
pers
onal
use
onl
y.
65 S. I. Tyritzis et al.
Ultrastruct Pathol
aCKnoWledGMent
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
referenCes
1. Renal cell carcinoma. European Association of Urology (EAU) Guidelines, 2010 edition.
2. Thompson HR, Frank I, Lohse MC, et al. The impact of isch-emia time during open nephron sparing surgery on soli-tary kidneys: a multi-institutional study. J Urol. 2007; 177: 471–476.
3. Μathieson PW. The cellular basis of albuminuria. Clin Sci. 2004; 107: 533–538.
4. Levidiotis V, Power DA. New insights into the molecular biology of the glomerular filtration barrier and associated disease. Nephrology. 2005; 10: 157–166.
5. Tyritzis SI, Kyroudi A, Liatsikos E, et al. Comparison of prolonged warm and cold ischemia on the solitary kidney during partial nephrectomy in a rabbit model. World J Urol. 2007; 25: 635–640.
6. K Evangelou, J Bramis, I Peros, et al. Electron microscopy evidence that cytoplasmic localization of the p16INK4A ‘‘nuclear’’ cyclindependent kinase inhibitor (CDKI) in tumor cells is specific and not an artifact: a study in non-small cell lung carcinomas. Biotech Histochem. 2004; 79: 5–10.
7. Asanuma K, Mundel P. The role of podocytes in glomerular pathobiology. Clin Exp Nephrol. 2003; 7: 255–259.
8. Shirato I. Podocyte process effacement in vivo. Microsc Res Tech. 2002; 57: 241–246.
9. Solez K, Racusen LC, Whelton A. Glomerular epithelial cell changes in early postischemic acute renal failure in rabbits and man. Am J Pathol. 1981; 103: 163–173.
10. Tveita T, Johansen K, Lien AH, Myklebust R, Lindal S. Morphologic changes in tubular cells from in situ kidneys fol-lowing experimental hypothermia and rewarming. APMIS. 2005; 113: 13–20.
11. Molitoris BA, Falk SA, Dahl RH. Ischemia-induced loss of epithelial polarity: role of tight junction. J Clin Invest.1989; 84: 1334–1339.
12. Andersson M, Nilsson U, Hjalmarsson C, Haraldsson B, Sörensson Nyström J. Mild renal ischemia–reperfusion reduces charge and size selectivity of the glomerular barrier. Am J Physiol Renal Physiol. 2007;292:1802-;1809
13. Rippe C, Rippe A, Larsson A, Asgeirsson D, Rippe B. Nature of glomerular capillary permeability changes following acute renal ischemia-reperfusion injury in rats. Am J Physiol Renal Physiol. 2006; 291: 1362–1368.
14. Sutton TA. Alteration of microvascular permeability in acute kidney injury. Microvasc Res. 2009; 77: 4–7.
15. Kwon O, Hong SM, Sutton TA, Temm CJ. Preservation of peri-tubular capillary endothelial integrity and increasing pericytes may be critical to recovery from postischemic acute kidney injury. Am J Physiol Renal Physiol. 2008; 295: 351–359.
16. Rouschop KMA, Roelofs JTH, Claessen N, et al. Protection against renal ischemia reperfusion injury by CD44 disrup-tion. J Am Soc Nephrol. 2005; 16: 2034–2043.
17. Johnston WH, Latta H. Glomerular mesangial and endothe-lial cell swelling following temporary renal ischemia and its role in the no-reflow phenomenon. Am J Pathol. 1977; 89: 153–166.
18. Terry BE, Jones DB, Mueller CB. Experimental ischemic renal arterial necrosis with resolution. Am J Pathol. 1970; 58: 69–83.
19. Alejandro V, Scandling JD Jr, Sibley RK, et al. Mechanisms of filtration failure during postischemic injury of the human kidney: a study of the reperfused renal allograft. J Clin Invest. 1995; 95: 820–831.
20. Konopka CL, Jurach A, Wender OC. Experimental model for the study of chronic renal ischemia in rats: morphologic, histological and ultra-structural analysis. Acta Cir Bras. 2007;22:12–21.
21. Mayrhofer C, Krieger S, Huttary N, et al. Alterations in fatty acid utilization and an impaired antioxidant defense mecha-nism are early events in podocyte injury: a proteomic analy-sis. Am J Pathol. 2009; 174:1191–1202.
22. Goujon JM, Hauet T, Menet E, Levillain P, Babin P, Carretier M. Histological evaluation of proximal tubule cell injury in isolated perfused pig kidneys exposed to cold ischemia. J Surg Res. 1999;82:228–233.
