A Comparative study of the protection offered by Vitamin E and captopril when used as additives in kidney storage solutions

The demand for organ transplantation has stimu­ lated a search for optimal methods of cadaveric kidney preservation techniques. The basis of de­ velopment of these methods depends on an un­ derstanding of cellular metabolic function during the period of storage and the changes confront­ ing the organ during reperfusion. Free radicals are implicated in both these phenomena. Conse­ quently agents that are capable of combating free radical attack (free radical scavengers/antioxidants) are now being investigated more exten­ sively, with the hope of providing better posttransplant organ function.


Introduction
Oxygen free radicals (OFR) have been implicated in tissue damage that occurs during cold storage and re-perfusion of donor organs (1,2). Recently developed methods of preserving organ viability during cold storage include incorporation of OFR scavengers into the storage solutions (3,4). For example, glutathione and glycine have been shown to protect liver tissue during cold ischaemia, and improve post-storage viability by OFR scavenging (4). In a previous study (5), we have also demonstrated that significant protection against changes in renal tissue metabolism (as assessed by the gluconeogenic potential) and ultrastructure, that occur during cold storage of rat kidneys in Marshall's citrate (MC) can be achieved by including vitamin E in the storage solution.
Angiotensin converting enzyme (ACE) inhibitors such as captopril that have thiol side groups have been reported to possess free radical scavenging properties (6) independent of their ACE inhibiting activity. The present investigation was carried out to compare the abilities of vitamin E and captopril when they are used as additives in kidney storage solutions, to protect rabbit kidneys against changes in renal function that occur during storage prior to transplant.

Preparation of kidney storage solutions (a) Marshall's Citrate solution (MC)
Marshall's Citrate solution was prepared accord ing to the method described by Ross el al (7). The final composition contained the following in Vitamin E was added to MC at a concentration of 50% of that of its LD 50, the dose that would result in the death of 50% of the animals under study.

(c) MC + Captopril
This solution contained MC to which captopril had been added at a dose of 10% of that of its LD 50. Since captopril is expensive, only 10% of its LD 50 was used in preparation of this solution.

Isolation of kidneys
Adult cross-bred albino rabbits (1.5 -2.0 kg body weight) were used in the study.
After induction of anesthesia by using an ether mask and injection of pentobarbitone sodium (0.5 mL/kg), the left kidney of each rabbit was exposed following a mid-line incision. The ureter was cleared and severed after leaving a length of about 5 cm from the kidney. The left renal artery and vein were ligated and divided and the kidney excised. The kidney weight ranged from 4-6 g. The kidney was immediately immersed in ice-cold MC, MC + El or MC + captopril solution. The renal artery was cannulated with a 23G butterfly needle and the vasculature was perfused with 50 ml of sterile solution of the respective group at 4° C, according to the solution tested. Fresh kidneys were used as controls. These were flushed with only MC. The right kidney was also removed in a similar manner and the kidney weight estimated. After flushing, the kidneys were stored for 24 hours at 0° C in ice, in the three different storage solutions that were being tested, and reperfused on the isolated perfusion kidney (IPK) circuit.

Reperfusion of kidneys in IPK model
The IPK model used in the present investigation was set up according to the technique described by Fuller and Pegg (9). The perfusate medium used was a modification of that used by Toffa ct al (10). The modifications concern a higher concen tration of creatinine and the replacement of hydroxyethyl starch by mannitol. The perfusate used in this study contained the following in mmol/mL: Na + 155, K + 6.8, Ca 2+ 2.0, CI" 139, HCCy 25, H 2 P0 4 1.5, glucose 5, creatinine. 0.2, and mannitol 100.0. Bovine serum albumin was included as a tracer for protein excretion assess ment, at a concentration of lg/L. The osmolality of the perfusate was 400 mosm/kg. Prior to use in the IPK machine, the above solution was fil tered through a series of filters and transferred to sterile containers. Control kidneys (freshly iso lated) were attached to the isolated perfusion cir cuit immediately after they were harvested. The other experimental kidneys were attached after they were flushed and cold stored for 24 hours in the respective storage solutions at 0° C in ice. The kidneys were subject to normothermic, blood less perfusion on this circuit for 30 -45 minutes after connecting to the IPK apparatus. Freshly prepared and filtered perfusate was then passed through. The perfusate was aspirated using a Watson-Marlow peristaltic pump (101 U: output 8 mL/min), from the reservoir and sent through a bubble trap to the renal arterial cannula. The arterial pressure was measured with a pressure transducer (Datascope 2002 A) situated on a sidearm close to the cannula.
The pressure was maintained between 100 and 120 mm Hg by the adjustment of the flow rate of the perfusate delivered to the kidney. The perfu sate medium in the reservoir was constantly, vig orously gassed with a mixture of 95% 0 2 :5% C0 2 . Samples of perfusate were collected at 15 minute intervals. Samples of urine were collected at 15 minute intervals. Samples of urine were collected at the end of every 15 minutes from the gradu ated tube that contains the ureter (i.e. samples were collected at 15 and 30 minute intervals). The first collection of urine was done after allowing the machine to run for about 10 minutes, as the urine flow may be irregular at the beginning of the cycle.

Assessment of renal function
During reperfusion, urine was collected and renal function assessed as described by Ross (7). The variables estimated for assessment of renal function were: (i) creatinine clearance (Ccr), (ii) tubular glucose reabsorption, (iii) tubular Na + reabsorption, (iv) urinary protein excretion. At the end of renal function assessment, the kidneys stored in the different storage solutions and fresh kidneys were examined by electron microscopy to detect any ultrastructural changes. Analysis of variance (ANOVA) was used when simultaneous comparisons were made with mea surements from more than two samples (12). The means between more than two samples were compared by ANOVA.
Students "t" test was used in the comparison of means when there were only two samples. A prob ability (p) value of <0.05 was taken to be statisti cally significant.
In this study where kidneys stored in different solutions (MC, MC+EI, MC + captopril) and the control group (fresh kidney) were analysed, the mean values of-the Ccr, sodium reabsorption, glu cose reabsorption, and protein excretion between each group was compared by the ANOVA. If the ANOVA showed that there was a significant dif ference of means between the groups (p<0.05), then the Tukey's Honestly Significant Difference (HSD) test was done (12). The calculated value for HSD signifies that in order to be statistically sig nificant at a probability of 0.05, the difference be-

Vol. 44 No.l, June 2001
tween any pair of means of the particular test must be at least equal to the HSD or more than it. This method also helps to determine which group has the best mean which does not differ significantly from the control mean.

Creatinine clearance
The results are shown in Table 1. Table 1 Creatinine

Tubular glucose reabsorption
The results are shown in Table 2. A decline in glucose reabsorption at 15 minutes of perfusion was observed in the kidneys during stor age as compared to the kidneys of the control group.
Of the stored kidneys glucose reabsorption was highest in the kidneys stored in MC+E1 (52.6 + 5.8). However a comparison of the mean glucose reabsorption values of the groups showed that there was no statistically significant difference between the means (p>0.05).

Tubular sodium reabsorption
The results are shown in Table 3. Table 3 Perecentage of sodium reabsorption of rabbit kidneys at 15 minutes of perfusion When the percentage mean sodium reabsorption values between the four groups (fresh, MC, MC+E1, and MC+captopril) were compared with each other (by ANOVA) there was a statistically significant difference between them (F=4.003, p < 0.05). Since there was a significant difference of the means between the groups as shown above, the Tukey's HSD test was done between the per centage mean sodium reabsorption of the four groups, to assess the best percentage mean so dium reabsorption close to the control group per centage mean. The HSD test showed that there was no statistically significant difference between the percentage mean sodium reabsorption of the control (fresh) group and the percentage mean sodium reabsorption of the kidneys stored in MC+E1 (p>0.05). This is to say the sodium reab sorption of kidneys stored in MC+E1 was next best to the results of the control (fresh) sample. However the HSD showed that there was a sta tistically significant difference between the per centage mean sodium reabsorption of fresh kidneys and kidneys stored in MC+captopril (p< 0.05). Results also showed that the difference between the percentage mean sodium reabsorp tion of kidneys stored in MC, was not significantly different (p>0.05) to those of fresh kidneys. These results show that the percentage mean sodium reabsorption values of kidneys stored in MC+ captopril were significantly less than that of fresh ones. The HSD test also shows that the difference of significance had occurred at the level of MC+ captopril and not with MC+E1 or MC.

Effects on urinary protein excretion
Urinary proteins were measured using commer cially available protein estimation kits (supplied by Randox laboratories Ltd.) using the biuret method (11) and expressed as a percentage of the perfusate protein concentration as done by Toffa el al (10). When the glomerular protein leakage was assayed by the above method it was found that protein leakage was high during the early period of perfusion (15 minutes). Kidneys stored for 24 hours in all the groups showed increased mean proteinuria when compared to the fresh (control) kidneys at 15 minutes of perfusion (Table  4). At 15 minutes of perfusion, the stored kidneys also showed a mean proteinuria which was higher than the perfusate concentration of protein, indi cating a net loss of protein. Results obtained indi cate that of the stored kidneys, the least amount of protein was excreted by kidneys stored in MC+E1 (Table 4). At 15 minutes of perfusion when the mean proteinuria values between the four groups (fresh, MC, MC+E1, and MC+captopril), were compared there was no statistically signifi cant difference between them (p>0.05).  Table 1.

Effects on kidney ultrastructure
Light microscopy did not show any morphologi cal changes in the kidney slices in any of the four groups that were examined (fresh, MC+E1, MC, and MC+captopril). There were no changes ei ther in the fresh (control) sample or on the samples of kidneys stored for 24 hours.
The glomeruli appeared normal and the lobular architecture was well maintained. Proximal tu bular cells were normal. The nuclei in these cells were clearly seen and the cytoplasm was eosinophilic. Interstitial oedema was not seen. No abnormalities could be detected on light micros copy in the proximal convoluted tubules.
However electron microscopy (EM) showed changes especially in the mitochondria of the proximal convoluted tubules which are active sites of glucose production. In fresh unstored kidney slices the mitochondria were normal and cylin drical shaped and sequentially arranged close to one another. The membranes of the mitochon dria were clearly demarcated and the cristae were seen clearly.
The electron microscopic appearance of the kid neys stored in MC+E1 for 24 hours showed very little change when compared to the control group. The mitochondria were cylindrical and arranged close to one another as in the fresh kidney. Most of the mitochondria were of normal size and shape. Mitochondrial membranes and the cris tae were clearly seen although some mitochon dria showed cellular swelling. In contrast, kid neys stored in MC and MC+captoril for 24 hours showed marked changes in the mitochondria of the proximal tubules when compared to fresh (control) ones. In the kidneys stored in MC the size and shape of the mitochondria were not uni form. The cylindrical shape was lost and the cells showed swelling. The cristae were not clear and different stages of damage to the mitochondria could be seen. The process of cellular disintegra tion had begun. The kidneys stored in MC+captopril showed the most extensive ultrastructural changes. Here the mitochondria had completely lost their normal architectural pat tern; they were hardly identifiable as mitochon dria. Gross mitochondrial swelling and disinte gration which leads to cellular destruction were seen. Cristae pattern was not seen at all. Thus the EM appearances of the kidney tissue stored in MC+E1 showed minimal changes when compared to the changes observed in those stored in MC and MC+captopril. The ultrastructure of the kidney was well preserved in kidneys stored in solutions containing vitamin El.

Discussion
Results of the present investigation demonstrates that in stored kidneys, the mean values obtained with respect to creatinine clearance, sodium reab sorption, glucose reabsorption, were lower than the corresponding values in the fresh control kid neys, while the values for urinary protein excre tion were higher than in the control kidneys. These results support the theory that there is a decrease in glomerular and tubular function in the kidney during ischaemia and reperfusion. These changes are probably due to the release of free radicals because a certain amount of protection against the above changes in renal function, could be achieved by including the free radical scavenger vitamin E into the kidney storage solution.

The Ceylon Journal of Medical Science
That vitamin E can help to protect against alter ations in glomerular filtration that occurs during ischaemia and reperfusion has also been demon strated by other researchers. An investigation with dog kidneys have shown that inclusion of vitamin E in Euro Collins flush storage solution results in a higher glomerular filtration rate (as sessed by the creatinine clearance) and a reduc tion in lipid peroxidation, compared to kidneys stored in Euro Collins solution containing no vi tamin E (13). It has also been demonstrated in rats, that a diet deficient in vitamin E can exacer bate the reperfusion injury due to the loss of oxi dant scavenging activity in the kidneys (14). Re sults of the present investigation demonstrate that inclusion of vitamin E in storage can help to pro tect against alterations in glomerular filtration and tubular reabsorption that occur in kidneys dur ing storage. Captopril, although considered to be a free radical scavenger (6), could offer little or no protection against renal function alterations in rabbit kidneys during storage and reperfusion. These results are also supported by observations in the morphological study.
There were only minimal changes in the electron microscopic appearance of kidneys stored for 24 hours in MC containing vitamin E as an additive, when compared to the control kidneys. The mi tochondria of the proximal tubules (which are the active gluconeogenic sites), show close similarity in appearance and arrangement, to those in the fresh kidneys. In contrast, in kidneys stored for 24 hours in MC or MC containing captopril, there were maked changes in the shape and size as well as loss of the normal architecture of tubular mito chondria. The kidneys stored in MC+ captopril showed the most extensive ultrastructural changes. Whether the differences in the protec tive effects of vitamin E and captopril were due to the differences in the amounts of these added to the storage solution is not clear. However, the fact that captopril is not an effective free radical scavenger is also demonstrated by results of other investigators (15,16). Results of the investigation by Westlin et al show that ability of captopril to protect ischaemic-perfused myocardium may be due to inhibition of compliment activation and other cardioprotective effects rather than due to free radical scavenging (15). Kukrije et al have shown that captopril is an ineffective superoxide radical scavenger (16). They also conclude that captopril, by virtue of its thiol (-SH) group, acts as a non-specific antioxidant.
From these results it can be concluded that vita min E.is a better free radical scavenger than captopril for use as an additive to flush-storage solutions, to protect against alterations in renal function that occur during ischaemia and reperfusion.