THE ROLE OF DIFFERENT CHEMICAL MODIFICATIONS OF SUPEROXIDE DISMUTASE IN PREVENTING A PROLONGED MUSCULAR ISCHEMIA/REPERFUSION INJURY

Abstract

It is well know that a long period of ischemia followed by reperfusion can create an irreversible tissue damage, also due to the excessive generation of oxygen-derived free radicals. A possibility for avoiding this syndrome is represented by the use of free radical scavengers, such as the superoxide dismutase (SOD). The current authors compared the results achieved through different modifications of this enzyme in an experimental rat hind limb model of ischemia/reperfusion. 60 rats that had a 4 hour and 30 minute ischemia of the left hind limb were divided into four groups of 15 each and treated using a physiological solution (control group), native SOD, monomethoxypolyethylene-glycol-SOD (mPEG–SOD) or poly(acryloilmorpholine)-SOD (PAcM–SOD). The outcomes obtained in terms of limb survival (p<0.05), as well as histomorphologic studies (p<0.0005), revealed a superior capacity of mPEG–SOD when compared with the other three substances.

INTRODUCTION

The ischemia-reperfusion (IR) syndrome can be related to tissue damage mediated by the excessive generation of oxygen-derived free radicals.[1-2] It is well known that among the factors that could influence the outcome of vascular, traumatic and reconstructive surgery, this syndrome has achieved a prominent position. In case of amputated limb replantation, free muscular flap transfer, Volkman's ischemia contracture, post-traumatic compartment syndrome, etc., the skeletal muscle is severely damaged and represents a great clinical problem. The large mass of ischemia-reperfused muscular tissue produced systemic and also local effects. In fact, after a long period of ischemia tissue reperfusion leads to the generation of highly reactive oxygen-derived free radicals, whose toxic effects amplify the tissue injury caused by ischemia.[3-6] The consequence is a whole of alterations, including endothelial swelling, microvascular thrombi and capillary leukocyte aggregation, which lead to the “no reflow” phenomenon and then to tissue death.[[7]]

Free radicals scavengers, in amounts exceeding the normal average may avoid this syndrome.

These substances affect the behaviour of the muscular tissue after an IR syndrome and were submitted to numerous experimental and clinical studies to establish their potential beneficial effects.[8-9] Among these enzymes, the most representative turned out to be catalase and superoxide dismutase.

The present paper describes the effects of a free radical scavenger, the superoxide dismutase, on the viability of the rat hind limb muscles after an IR injury: the native enzyme and two different chemically modified forms were studied and immediately administered after a prolonged ischemic period following by reperfusion.

MATERIALS AND METHODS

Native and mPEG–SOD

The native enzyme superoxide dismutase (SOD) was obtained from the Diagnostic Data Incorporation (Mountain View, CA, USA) and possessed a specific activity of 4500±380 U/mg. To conjugate the polymer with the enzyme, monomethoxypolyethylene-glycol (mPEG) was activated as succinimydil ester withnor-leucine as spacer following the method reported by Sartore et al.[10-11] The final conjugate was found to present 63% of the starting enzyme activity and about 11 mPEG chains bound to each enzyme molecule. The approximate molecular weight of the conjugate can be assumed to be 87,000 Da. The conjugate was recovered by lyophilization into various vials and maintained in the dried state at 0°C. The enzyme-polymer conjugate solution was reconstituted shortly before use.

PAcM–SOD

Carboxy terminated poly(acryloilmorpholine) (PacM) was obtained by radical polymerisation of acryloilmorpholine using azodiisorbutyronitrile as initiator and 2-mercaptoacetic acid as chain terminator. The polymer was fractionated by solvent precipitation through portionwise addition of ethyl volumes to a polymer solution in an isopropylic alcohol and methylene chloride mixture 7:3 v/v. The fraction corresponding to a mol/1 wt of 6000 D was activated as succinimydil ester for protein binding.[[12]]

One hundred and twenty seven mg of activated PacM were added under vigorous stirring to 10 mg of bovine liver SOD (DDI, Mountain View, CA, USA) and dissolved in 1 ml of 0.2 M borate buffer pH 8.0, in order to reach a protein amino group/polymer in the molar ratio of 1:3. The reaction solution was maintained at room temperature under stirring for 1 hr and the protein-polymer conjugate (PacM–SOD) was purified by gel filtration chromatography using a preparative Superose 12 column eluted with 10 mM phosphate buffer, 0.15 M NaCl, pH 7.2, operated on a FPLC system. The eluted fractions were analysed by OD at 280 nm for protein detection, by iodine test[[13]] for polymer determination and by enzyme assay.[[14]] The protein concentration of the collected PacM–SOD purified fraction was estimated by biuret.[[15]] The degree of modification (percentage of polymer derived protein amino groups) was estimated by the Habeeb method (1966)[[16]] to be 46% value that corresponds to a mean of 10 polymer chains per protein molecule. The enzyme activity was found to be 60% of the unmodified form (native SOD).

Surgical Procedure

Under general anesthesia, 60 Wistar rats (350±50 g b.w.) were submitted to a 4 hr and 30 min ischemia of the left hind limb, after its microsurgical division from the thigh and clamping of the femoral artery as described in a previous paper.[[17]] The animals were divided into four groups of fifteen each; the first ones were considered as a control group (no therapy: injection in the epigastric vein of 2 cc of saline solution only), while the remaining three were treated at the end of ischemia with an injection in the epigastric vein of 15,000 U/kg of native SOD, 3000 U/kg of mPEG–SOD and 15,000 U/kg of PacM–SOD, respectively, diluted in 2 cc of saline solution. All of the experiments were conducted in the same operating theatre by the same operator and in the same environmental conditions (T=24±0.5°C; Relative Humidity=55%±5%). Saline soaked gauze was placed on and around the muscles of the limb to prevent desiccation. During the total ischemic period, the temperature of the legs was 29±0.5°C. All animals received food and water ad libitum before and after surgery. No antibiotic therapy was administered. A further group of three rats was included in the experiment: they underwent femoral artery ligature without any therapy.

Monitoring

The microcirculation of the operated limb was monitored on the sole by a Laser Doppler Flowmeter (PF3—Periflux—Perimed, Sweden), using a needle probe (Periflux 303) measuring 1 mm in diameter. The perfusion monitor was calibrated with a 250 motility unit reference standard. Since the probe and the fiberoptic cable are extremely sensitive to motion artefacts or to changes in contact pressures, a micromanipulator was employed to maintain the probe perpendicular to the tissue.

These measurements were taken before ischemia (steady state), during ischemia, at the beginning of reperfusion and one hour after reperfusion. Once a stable signal was obtained, each measurement lasted one minute and was then repeated in five different adjacent sites, since the Laser Doppler Flowmeter revealed marked spatial variations.[[18]] Flux values were expressed in perfusion units (PU) as recommended by the manufacturer (1 PU=10 mV), and when they ranged from 0 to 1 PU, they were considered as 0 PU.[19-20] The results were indicated as the percentage difference between basal values and results achieved.

In addition, other data were continuously recorded, such as mean arterial blood pressure (MABP), heart rate (HR) and body temperature (T), by means of an integrated modular system (HP M1165A, Model 56S, Hewlett Packard, Medical Products Group, Andover, MA, USA).

Histology

Muscle biopsy from the flexor digitorum superficialis were taken after one hour of reperfusion and then on the 4th, 6th, and 10th day. Control biopsies were performed at the same level in controlateral normal limbs. Each sample was quickly frozen in cooled isopentane (−160°C) and stored at −90°C, until at all of the samples wre processed as a single batch. Eight μm-thick transverse sections were cut from each sample using a cryotome (−20°C), mounted on slides and stained using hemotoxylin and eosin (HE), as well as modified Gomori trichrome. In addition, the sections were stained with the following histochemical reactions: NADH-tetrazolium reductase (NADH–TR), succcinate dehydrogenase (SDH), acid phosphatase (naphtol AS phosphate method), alkaline phosphatase, non-specific esterase and adenosine triphosphatase (ATPase) (Bio-Optica, Milan, Italy).

The method selected to measure the size of the muscular fibers was the “lesser fiber diameter” technique, defined as the maximum diameter across the lesser aspect of the muscle fiber.[[21]] The histomorphologic analysis of the muscular fibers was then carried out on all the limbs that had survived 10 days, employing the Imaging System KS300 Rel. 2.0 (Kontron Elektronik GmbH, Eching bei Munchen, Germany). The same software was used to count the muscular fibers which presented central located nuclei in all the limbs that had survived ten days and in ten normal limbs.

Statistical Analysis

Statistical analysis was carried out using the software package SPSS/PC+Statistics™ 7.5 (SPSS, Chicago, IL USA). Data are reported as mean±standard deviations at a significance level of p<0.05. Chi-squared (χ2) and one-way ANOVA were used to compare survival/perfusion data and histomorphometric findings, respectively, among the different group.

Animal care complied with the 86/609 EEC directive and Italian law (D.L. 116/92) for laboratory animals.

RESULTS

During the different surgical phases, mean HR, MABP and T were 223.1±15.3 beats/min, 66.5±11.4 mmHg and 34.2±0.5°C, respectively. These values did not change significantly during surgery, except for the moment of the anaesthetic injection; moreover, they did not reveal any significant inter-group differences (data not shown).

All animals tolerated surgery and ischemia/reperfusion time well, and they survived the initial ischemic insult. All the operated limbs showed a post-ischemic paresis. An edema of the operated leg, which disappeared on post-operative day 3/4, was evident in all the survived limbs. After the ischemia/reperfusion insult, the survival percentage of the operated limbs for the four groups was 33%, 60%, 87% and 73%, respectively (χ²=10.05; p<0.05). The three legs where the femoral artery was ligated, became necrotic in two days.

The average perfusion values recorded on the sole of all the animals were following steady 112.84±42,6, ischemia 5±5, reperfusion 41.68±31.55, 1 hour after reperfusion 86.03±63.05.

In particular, a range of predictive (95%) perfusion values for limb healing and necrosis was identified at the beginning of the reperfusion and turned out to be 87–108 PU and 34–70 PU, respectively. A one-way ANOVA, performed in all of the four groups and on the percentage of variation between data collected at the beginning of the reperfusion and data registered during the steady state, showed significant differences as reported in Table 1.

Table 1. Percentages of the Sole Flux Variations in the Four Groups at the Beginning of the Reperfusion (Mean±SD, n=15)

Animal Groups	F% variation
Control	−82±19.6^*
SOD	−44.6±6.5°
MPEG–SOD	−59.1±8.9
PacM–SOD	−57.7±15.1

ANOVA F=17.01 p<0.005.

SOD: superoxide dismutase; mPEG–SOD: monomethoxypolyethylene-glycol-SOD; PacM–SOD: poly-(acryloilmorpholine)-SOD.

Scheffé's post hoc multiple comparison test: ^*Control vs. mPEG–SOD (p<0.001), SOD and PacM–SOD (p<0.0005); °SOD vs. mPEG–SOD (p<0.05).

The results of the histomorphologic analysis confirmed that in vital limbs was a trend towards muscular fiber healing in the postoperative period; that was demonstrated by a positive acid phosphatase staining of the period, followed by an increase in the positive alkaline phosphatase. After ten days, the histomorphologic analysis comparing control group, normal limbs and the three groups of treated limbs (Table 2) revealed significant data in terms of fiber diameter (F=22.91; p<0.0005) and percentage of central located nuclei (F=93.34; p<0.0005).

Table 2. Histomorphologic Data of the Vital Limbs (Mean±SD)

Group	n.	Fiber Diameter (μm)	% Fibers with Central Located Nuclei
Normal limbs	10	75.6±4.8	2.6±0.9
Control	5	67.2±6.8	28.7±4.7
SOD	9	49.2±10.9^**	45.9±14.8^#
MPEG–SOD	13	43.1±6.8°	87.4±5.5°°
PacM–SOD	11	53.3±12.3^*	46.9±16.4^#

ANOVA F=22.91; p<0.0005; F=93.34; p<0.0005.

SOD: superoxide dismutase; mPEG–SOD: monomethoxypolyethylene-glycol-SOD; PacM–SOD: poly-(acryloilmorpholine)-SOD.

Scheffé's post hoc multiple comparison test:

- Fiber diameter: ^*PacM–SOD vs. Normal limbs (p<0.0005); °mPEG–SOD vs. Control and Normal limbs (p<0.0005); ^**SOD vs. Control (p<0.05) and Normal limbs (p<0.0005).

- % Fibers with central located nuclei: °°mPEG–SOD vs. Normal limbs, Control, SOD and PacM–SOD (p<0.0005), ^#SOD and PacM–SOD vs. Normal limbs (p<0.0005).

DISCUSSION AND CONCLUSION

According to the available literature, there is a great variability in the time interval required to create a severe, but potentially reversible, muscular necrosis subsequent to a complete ischemia without collateral flow, and that depends on the animal species selected; to provide some examples, we can mention about four hours in the rabbit rectus femoris,[[8]] five hours in the canine gracilis[[22]] and three hours in the rat hind limb tourniquet model.[[23]] In all of the three models listed above, a 6–7 hour ischemia causes a total and irreversible muscular necrosis. With respect to this issue, we found our period of four hours and thirty minutes more than sufficient to create a severe ischemia/reperfusion damage to the rat hind limb.

The beginning of the reperfusion is the critical point at which free radical antagonists have to show their effectiveness. The aim of the present study was to determine whether two different chemical modifications of the same free radical scavenger could improve the skeletal muscle protection against the IR.

After a prolonged ischemia, tissue reperfusion with oxygen-enriched blood causes a rapid breakdown of the hypoxanthine stored up during the ischemic phase. This reaction is responsible for an enormous production of xanthine and free superoxide radicals, followed by tissue damage.[[7]] Since SOD terminates superoxide ions, it is predictable that it could enhance muscular tissue survival after a prolonged ischemia. This survival time could be further improved by means of chemical modifications that could increase the native SOD normal capacity. In our study, such a result was obtained by prolonging the Sod plasmatic half-life (mPEG–SOD and PacM–SOD), or by decreasing its antigenicity and at the same time increasing its percentage in the endothelial and red blood cells (mPEG–SOD) with respect to native SOD.[[24]]

To ensure the highest accuracy when determining the degree of ischemia reached in the present experiments, the current authors adopted an LDF monitoring system, its accuracy in this field being already demonstrated.[[18]] In fact, in a previous study[[25]] sole flux measurements taken by means of the Laser Doppler Flowmeter (LDF) technique, had been found to be reliable when evaluating both degree and efficacy of the induced ischaemia. Subsequently, LDF measurements were performed only in this site and not in the muscles.

According to the results obtained, the mPEG modification was clearly proven to be the best one of the three tested enzymes, and several points supported this conclusion. First of all, it should be taken into account that the enzyme dose administered in our protocol was different: the mPEG–SOD quantity was five times lower than native SOD and PacM–SOD (3000 U/kg compared to 15,000 U/kg). The reason for that is that the chemical and structural modifications supported by native SOD improve its biological half-life by a few minutes up to 24 hours in case of mPEG–SOD, or up to 6 hours in case of the PacM modified form.[[24]] Therefore, a lower dosage was administered for the longer plasmatic half-life form, in order to stress the validity of the selected model.

Then, histomorphologic results turned out to be better after using mPEG–SOD (Table 2. The mean finer diameter and percentage of fibers with central located nuclei were the two parameters selected, since after necrosis regeneration can occur in continuity with the undamaged portions of the fiber (continuous regeneration). In normal muscles, up to 3% of the fibers may contain central nuclei; furthermore, regenerating fibers often develop in small groups and, even if single, they are usually smaller than normal fibers.[[26]]

In conclusion, the present outcomes strongly indicate that when preventing ischemia/reperfusion injury, SOD modified with mPEG is more effective than the other chemical modifications of the same enzyme.

ACKNOWLEDGMENTS

Financial support for this research was partially given by Fondazione Cassa di Risparmio of Bologna, the Rizzoli Orthopaedic Institute and the University of Bologna, Bologna, Italy.

We thank Patrizia Nini, Franca Rambaldi, Patrizio Di Denia and Claudio Dal Fiume (Department of Experimental Surgery) for their technical assistance.