Superoxide dismutaseRADIOBIOLOGICAL SIGNIFICANCE. OCCURRENCE IN HUMAN TISSUES,

TUMOURS AND TUMOUR CELL-LINES.

AKADEMISK AVHANDLINGsom med vederbörligt tillstånd av rektorsämbetet vid Umeå Universitet för avläggande av medicine doktorsexamen kommer att offentligen för­svaras i Föreläsningssalen, 2 tr., Institutionen för Onkologi (Jubi- leumskliniken), Umeå Universitet, fredagen 14 oktober 1983 kl. 09.00

av

Gunnar Westmanmed.lic.

UMEÀ 1983

ABSTRACTSuperoxide dismutases (SOD) are enzymes found in all aerotolerant or­ganisms, catalyzing the dismutation of superoxide radicals, in the fastest enzyme-catalyzed reaction known:

2 H+ + 2 0 “ ’ -------> 0 2 + H2 0 2

Since superoxide radicals are formed in large amounts during aerobic ir­radiation, an experimental investigation on the possible significance of SOD in radiobiology and radiotherapy was of interest.

In the thesis the concept was studied in two different ways:

I. The activities of SOD (and for comparison also the activities of the hydrogen peroxide degrading enzymes catalase and glutathione peroxi­dase) were determined in human tissues, tumour and normal and neo­plastic human cell-lines. The investigation gives comprehensive data on the enzyme activities in these systems and shows that CuZn SOD and glutathione peroxidase activities are evenly distributed, while Mn SOD and catalase activities vary considerably between different tissues, tu­mours and cells. However, no obvious relation was found between the enzyme activities and empirically assumed radiosensitivities.

II. A procedure was developed for inhibition of CuZn SOD in cells in vitro, by the chelating agent diethyldithiocarbamate, (DDTC). Tempo­rary inhibition of CuZn SOD activity at the time of irradiation or in the first post-irradiation hours was shown to increase radiation damage to Chinese hamster cells, whilst no increase was seen for radiation-induced haemolysis of erythrocytes.

It is concluded that inhibition of CuZn SOD in actively dividing and me­tabolizing cells increases radiation damage probably through mecha­nisms related to potential lethal damage.

Key words: Superoxide dismutase, Radiation sensitivity, Diethyldithio­carbamate, Haemolysis, Catalase, Glutathione Peroxidase, Potential lethal damage.

UMEÀ UNIVERSITY MEDICAL DISSERTATIONS New Series No 94 ISSN 0346-6612

Umeå University Medical Dissertations New Series

No 94From the Department of Oncology University of Umeå, Umeå, Sweden

SUPEROXIDE DISMUTASE

RADIOBIOLOGICAL SIGNIFICANCE. OCCURRENCE IN HUMAN TISSUES,

TUMOURS AND TUMOUR CELL-LINES.

byGunnar Westman

- 2-

This dissertation is based on the following papers referred to in the tex t bytheir Roman numerals:

I Westman, Gunnar and Marklund, Stefan L.; Diethyldithiocarbamate, a Superoxide Dismutase Inhibitor, Decreases the Radioresistance of Chinese Hamster Cells. Radiat. Res. 83, 303-311 (1980).

II Westman, N. Gunnar and Marklund, Stefan L.: Copper- and Zinc- containing Superoxide Dismutase and Manganese-containing Super­oxide Dismutase in Human Tissues and Human Malignant Tumors. Cancer Res. 41: 2962-2966, 1981.

III Marklund, Stefan L., Westman, N. Gunnar, Lundgren, Erik, and Roos,

Göran : Copper- and Zinc-containing Superoxide Dismutase, Manga­nese-containing Superoxide Dismutase, Catalase, and Glutathione Peroxidase in Normal and Neoplastic Human Cell Lines and Normal Human Tissues. Cancer Res. 42: 1953-1961, 1982.

IV Westman, N. Gunnar, and Marklund, Stefan L .:‘ Diethyldithiocarba­mate inhibition of CuZn superoxide dismutase in human erythrocytes: no increase in radiation haemolysis. Int. J. Radiat. Biol. 43: 103-109, 1983.

V Westman, Gunnar and Midander, Jörgen: Post-irradiation diethyl- dithiocarbamate-inhibition of CuZn superoxide dismutase reduces clonogenic survival of Chinese hamster V 79 cells. Accepted for publication in Int. J. Radiat. Biol.

- 3-

CONTENTS

1 GENERAL BACKGROUND 31.1 Introduction 31.2 Free radical chemistry 61.3 Intermediates in oxygen reduction 71.4 Enzymatic defense mechanisms 101.3 Nonenzymatic defense mechanisms 131.6 Lipid peroxidation 131.7 Radiation and oxygen free radicals 161.8 Diethyldithiocarbamate (DDTC) 18

2 PRESENT INVESTIGATION 23

2.1 Aims 232.2 Materials and Methods 242.3 Results and Discussion 382.3.1 Radiation experiments 382.3.2 Superoxide dismutase in tissues, tumours and cell-lines 43

3 CONCLUSIONS 33

4 ACKNOWLEDGEMENTS 34

5 REFERENCES 33

- 5-

1 GENERAL BACKGROUND

1.1. Introduction

Oxygen is absolutely essential for the life of higher organisms. Never­theless life on earth began in the abscence of oxygen, in a reducing atmosphere. The appearance of the first blue-green algae approximately 2 billion years ago and the subsequent oxygenation of the biosphere through photosynthesis, imposed a strong evolutionary pressure on many organisms tha t up to then had lived and evolved in the world. The presence of oxygen was dangerous for most compounds containing carbon, hydrogen and nitrogen, which in the presence of oxygen easily oxidize to the correspond­

ing oxides, C to CC>2 , N to NO2 and H to H2O. In parallel to developing mechanisms for the utilization of oxygen for energy production the organisms also had to to develop defenses against the toxicity of oxygen.

However, oxygen is still toxic to most organisms a t concentrations that are only slightly greater than those in normal air. Oxygen toxicity is seen in whole animals, plants and bacteria, and also in cell culture, homogenates and isolated organelles. Breathing pure oxygen rapidly damages lungs and

causes retinal damage in newborn babies. D ifferent aspects of oxygen toxicity has, therefore, a ttrac ted considerable in terest since more than a century (19,55,81).

Oxygen toxicity is intim ately related to the generation of free radicals, and the discovery by McCord and Fridovich (109) in 1969 of the enzyme superoxide dismutase (SOD), has opened new horizons for the research on oxygen toxicity. SOD is the first enzyme found which catalyzes a reaction between free radicals, the superoxide radicals, which are the first uni­valent reduction products of molecular oxygen.

Free radicals are of great scientific in terest in a broad field of biology. In human biology and pathology they are associated with ageing (79), inflammatory processes (111), autoimmune diseases (49), ischemic diseases (132), and with neoplastic diseases (107,108,166,167); with the effects of

- 6-

radiation (31), phagocytic killing of microorganism (10), killing of tumour cells by NK cells (144) and with the action and toxicity of several drugs and chemicals (11,33,68,141,132,164).

Interm ediates of oxygen reduction, therefore, are common in the cell and consequently knowledge of our defenses against these deleterious species is an im portant field in scientific research. The present study contribute to this field with special emphasis on the occurrence of protective enzymes in tissues and cells and the possible significance of SOD in radiobiology and radiotherapy.

1.2. Free radical chemistry

A free radical (R*) is defined as any atom or molecule with one unpaired

electron occupying an outer orbital. A radical can be charged, e.g. H3N + (protonated amine radical), (superoxide anion radical, superoxide radical, superoxide), or uncharged, e.g. NO (nitrous oxide). Free radicals are generally very reactive; some like the superoxide radical being considered less dangerous than others, like for example the hydroxyl radical (OH*). Free radical reactivity is accounted for by the strong tendency of the unpaired electron to in teract with other electrons to form an electron pair. Radical reactions are usually divided into three types, initiation, propagation and term ination reactions.

Initiation means the dissociation of a covalent bond,

R:R ---------------- + R*

and may be accomplished by therm al energy, electrom agnetic energy (e.g. radiation), or by energy derived from oxidation-reduction coupling with a source of free radicals (e.g. H2 O2).

a) therm al decomposition: H2O2 --------- ^ 2 OH*

b) radiolysis of water: H2O OH*,e”,H*, H2 O2 ..........

- 7-

c) redoxcoupling with Fe: Fe2+ + H2O2 --------- >*Fe3+ + OH” + OH*

Reaction rates are modified by the products formed - (i.e. the effec t of resonance stabilisation), heat of formation and presence of catalyzing compounds.

A fter formation of free radicals they may be propagated by several mechanisms, e.g. addition and tranfer reactions:

d) atom transfer reaction: R* + R 'H --------- >*RH + R1*

e) addition reaction: R* + -C=C-----------► R -C -C -

Propagation finally ends with term ination reactions, e.g. combination or disproportionation reactions:

f) combination reaction: R* + R’ -------- ►R-R

g) disproportionation reaction: R* + R*CH2 -C -H R "--------- >*RH + R'CH=CHR"

1.3. Interm ediates in oxygen reduction

1.3.1. Molecular oxygen, O7

Molecular oxygen in its ground sta te is a biradical with two unpaired electrons with parallel spins in its outer orbital. This parallel electron spin arrangem ent prevents the direct addition of a pair of electrons with antiparallel spins, and an unfavourable so-called spin inversion is necessary for this process to occur. Spin inversion is a slow process and so there is seldom enough tim e for it to occur, during the life-tim e of the collisional complex. Therefore, if possible the one electron reduction of oxygen, involving no spin inversion, would dominate over the two electron reduc­tion, (162).

- 8-

_ e- n - e- + 2H+ H n e’ + H* nu. e" +H+0 2 — ► 0 2 ------------------- ► h2 ° 2 -H^O H20

Figure i . One-electron reduction of molecular oxygen.

1.3.2. Superoxide radical, 0 \

The first step in the univalent reduction of oxygen results in formation of the superoxide radical. Superoxide is the conjugate base of a weak acid

HO^, with a pKa of about 4.8. Thus a t the pH of the living cell about 7.4, the superoxide radical dominates over the perhydroxyl radical by almost a factor of 103 (18).

Superoxide acts both as a reducing agent and as an oxidizing agent, i.e. donates or accepts electrons. In general the reactivity is low compared to other free radical species, and the damaging effects of superoxide there­fore must be accounted for by the secondary production of other, more reactive species. The low reactivity on the other hand permits a long life­tim e and a large diffusion capacity. The superoxide radicals can, therefore, be considered to play the role of a transm ittor or generator of 'site- specific1 hydroxyl radicals (24,50).

Superoxide radicals are generated in biological systems in many different ways. Several enzymes produce during oxidation-reduction reactions (e.g. xanthine oxidase, aldehyde dehydrogenase, dihydroorotic dehydro­genase and several flavin dehydrogenases (55). Drugs and chemical compounds produce superoxide radicals on their autoxidation or m eta- bolisation in the cell. Hemoglobin and myoglobin in their oxygenated form slowly liberate superoxide radicals when they are transformed to methemoglobin and metmyoglobin (55). In the cell superoxide generation has been dem onstrated in fragments of organelles such as mitochondria (42), microsomes (8 ), chloroplasts (130) and also in nuclear prep­arations (14).

- 9-

The concentration of superoxide radicals in the aerobic cell is maintained a t about 10“ 11 M (168). Most of the radicals are formed in mitochondria, and the largest generation of radicals take place in the liver. On the whole, probably only a few percent of the to ta l amount of oxygen consumed by the cell is converted to superoxide radicals (55). This is nevertheless a considerable quantity of radicals, due to the large amount of oxygen metabolized per unit tim e in the cell.

1.3.3. Hydrogen peroxide, H9O9

Hydrogen peroxide is generated when oxygen is reduced by two electrons (divalent reduction) or when superoxide is reduced by one electron (univalent reduction). Hydrogen peroxide in itself seems to be quite unreactive in the cell (301). This is due to two factors: i) most of the peroxide is generated in small special subcellular organelles called peroxi­

somes, where the hydrogen peroxide degrading enzyme catalase is also localized, ii) the low concentration in other parts of the cell (10 “ 8 M (130). High concentrations of hydrogen peroxide on the other hand is cytotoxic.

Hydrogen peroxide is produced by several enzymes: for example urate oxidase, L-amino acid oxidase and glycollate oxidase (in the peroxisomes) which transfer two electrons from their substrates onto each oxygen molecule. Most of the hydrogen peroxide in the cell is, however, produced by univalent reduction of superoxide, e.g. by the enzyme, xanthine oxidase.

1.3.4. Hydroxyl radical, OH*

The hydroxyl radical is generated by univalent reduction of hydrogen peroxide, or directly from water by ionizing radiation. This radical is very reactive and a very strong oxidizing agent (30). It removes electrons from other molecules and reacts with almost every other compound in a very short time. In view of its high reactivity (k=10^ to 10H ), the existence of free hydroxyl radicals in the cell has been questioned (24).

- 10-

1.3.5. Transition metal catalysis

The presence of transition metals, e.g. iron, in the cell, is potentially dangerous as iron may catalyse several types of toxic redox reactions. Iron within the cell is not ”free" but is chelated with ubiquitous cellular constituents as amino acids or specific proteins. Iron chelates are more soluble than free ferrous iron a t or near neutral pH. Ferrous ions chelated with ADP, phosphate or other chelating substances undergo rapid autoxida- tion with generation of both hydrogen peroxide and hydroxyl radicals (36,75). Ferric iron complexes react rapidly with superoxide, which by donating its electron to the iron reduces it to the ferrous form and in the presence of hydrogen peroxide, hydroxyl radicals are formed. This reaction has been called the iron-catalyzed Haber-Weiss reaction and is of central importance in the discussion of the dangers of superoxide in biology (78).

0~2+ F e 3 + ------------ * * 0 2 + F e 2+

Fe2+ + H20 2 --------- **Fe3+ + OH- + OH*

Figure 2. The iron-catalyzed Haber-Weiss reaction.

1*̂ * Enzymatic defense mechanisms

The oxygen interm ediates are too reactive to be tolerated by living organisms. Considering the common evolutionary pressure on a varied biota it is not surprising that multiple defense mechanisms evolved, (figure 3). However, in anaerobic organisms these mechanisms are often absent or may exist a t an insignificant level (110).

C A TALASEPE R O X ID A SE

► OH ► H20

SUPEROXIDE D ISM U TA SE

Figure 3. Enzymatic defense mechanisms to bypass and prevent the accumulation of reactive oxygen interm ediates.

- 11-

1.4.1. Superoxide dismutase

Superoxide dismutase, was first isolated as a 35,000 molecular weight copper protein in erythrocytes and liver by Mann and Keilin in 1939 (101). In the mid-1960 the protein was isolated (without knowledge of its dismutating function) as an antiinflammatory protein by Huber e t al. (87). An enzymic function of the enzyme was discovered in 1969 by I. Fridovich and 3. McCord a t Duke University (109). The enzyme catalyzes the disproportionation of superoxide radicals:

2H+ + 20_2---- ^ 0 2 + H20 2

Thus the reaction occurs in a sequential pathway involving consecutive reactions in which two identical substrate molecules dism utate to higher

and lower oxidation sta tes. The reaction is the fastest enzyme reaction known with a reaction ra te constant (k) for the CuZn form of the enzyme of:

k = 2 - it X 109 M -ls-la t pH 7 (55)

The enzyme works efficiently even a t the low intracellular enzyme concentrations.

Several types of the enzyme exist: in the eukaryotic cytosol there is a

CuZn SOD with a molecular weight of 33,000, and in mitochondria a Mn SOD with a molecular weight of about 85,000. Recently a new extra­

cellular CuZn containing enzyme was isolated from human lungs, the enzyme being distinctly different from the ordinary CuZn SOD with a molecular weight of about 135,000 (105). Prokaryotic organisms have both a Mn SOD and a Fe SOD. The CuZn SOD and Mn SOD are readily distinguished, even in crude exracts, since the former is inhibited by cyanide and is resistant to treatm ent by a mixture of chloroform and ethanol, whereas the la tte r is resistant to cyanide but is denatured by chloroform and ethanol. Hydrogen peroxide inactivates CuZn SOD in vitro

a t alkaline pH, through oxidative a ttack on a histidine residue in the active site of the enzyme (85).

- 12-

Superoxide dismutases have been found in all aerotolerant organisms but not in obligate anaerobic organisms (110). The manganoenzyme is induced by oxygen in microorganisms and is also inducible in mammalian cells and tissues. The mitochondrial and bacterial Mn SOD have homologous amino acid sequences (153) in agreem ent with the hypothesis tha t mitochondria have evolved from an endocellular symbiosis between a prokaryote and a protoeukaryote.

It may be questioned as to why an enzyme is needed for the catalysis of a reaction tha t is so rapid, even in the abscence of catalysis. As Fridovich (55) has discussed, however, the presence of 1x10“^ M CuZn SOD in the cell increases the rate of the dismutation of 1x10” 10 M superoxide by a factor of 109.

1.4.2. C atalase

C atalase (see review 32), is a hemoprotein (i.e. it contains iron), and its existence has been known since the beginning of the 19th century. It is present in virtually all mammalian cell types and catalyze the dispropor­tionation of hydrogen peroxide to oxygen and water:

2 H 2 O 2 ------------ ^ 0 2 + 2H 2O

C atalase works equally well a t high and low concentrations of oxygen but can not utilize other peroxides than hydrogen peroxides. The enzyme is localized in subcellular organelles such as the peroxisomes (microbodies) of the liver and the kidney and in other cells in much smaller aggregates such as the microperoxisomes and sometimes in the mitochondria.

1.4.3. Glutathione peroxidase

Glutathione peroxidase (see review 53), was discovered in 1957 by Mills (116), but the presence of selenium in the active site of the enzyme was not reported until 1972 (52). This enzyme catalyzes the reaction between several types of hydroperoxides (apart from hydrogen peroxide) and requires reduced glutathione (GSH) as a hydrogen donator. Also lipid

- 13-

hydroperoxides may be inactivated by the enzyme, preventing thereby the propagation of the radical chain reactions during lipid peroxidation. During the reduction of the peroxide, GSH is oxidized to GSSG (oxidized gluta­thione) The reduction of GSSG back to GSH is accomplished by the enzyme glutathione reductase using NADPH as hydrogen donator. The NADPH is then regenerated by the hexose monophosphate pathway (HMP shunt).

GSHPX2 GSH + ROOH ------------ ► GSSG + ROH + H20

GSHREDGSSG + NAD(P)H + H+ ------------► 2 GSH + NAD(P)+

Glutathione peroxidase works well a t low concentrations of hydrogen peroxide but becomes saturated a t higher concentrations. The subcellular distribution of glutathione peroxidase is complementary to th a t of catalase, with two thirds of the enzyme in the cytosol and one third in the mitochondria (32). The relative contribution of catalase and glutathione peroxidase to hydrogen peroxide degradation probably varies from cell type to cell type.

1.5. Nonenzymatic defensive mechanisms

Probably the most im portant hydrophilic defensive mechanism in the cell is reduced glutathione (GSH), which is a small tripeptide (y-Glu-Cys-Glu). Glutathione (see review 91) has several functions in the cell. In connection with protection against free radical damage it is able to: i) scavenge

hydroxyl radicals, ii) donate its hydrogen to radical-damaged target molecules, and iii) function as hydrogen donor for the action of glutathione peroxidase.

The normal concentration of GSH in the cell is quite high 1-50 x 10"^ M.

As oxidized glutathione GSSG, is very toxic to the cell (through mixed disulfide formation) the cell tries to maintain the ratio GSH/GSSG a t a constant level. Other thiol molecules in the cell as cysteine and protein

- 14-

bound sulfhydryls also contribute, although to a minor extent, to the defensive action of intracellular sulfhydryls.

Ascorbic acid reacts rapidly with both superoxide and hydroxyl radicals. It has been found to be radioprotective in vitro (129) and in vivo (2), but like oxidized GSH, ascorbic acid that has donated its hydrogen, may also be deleterious since it combines with or damages cellular constituents, especially if traces of metal ions are present. Therefore animal cells possess a NADH semidehydro ascorbate reductase that presumably functions in vivo to regenerate ascorbate from its oxidation products (61).

Compounds that form complexes with transition metal ions, as transferrin, caeruloplasmin, ferritin and albumin are also of importance, as they may prevent m etal-catalyzed hydroxyl radical generation. Artifically added simple aliphatic alcohols, thiourea and other thiols may also scavenge hydroxyl radicals.

The hydrophobic regions of the cell are the lipid membranes with their content of polyunsaturated fatty acids (PUFA). Since oxygen is 7-8 times more soluble in non-polar media and metal complexes such as heme are present in membranes, membranes are prone to undergo lipid peroxidation (vide infra) with subsequent structural damage to the cell. The major defensive mechanisms against this peroxidation are vitamin E (a-toco- pherol) and vitamin A (a retinoid, retinol), and the enzyme glutathione- peroxidase.

Vitamin E (see review 176), functions by interfering with the propagation

of radical chain reactions in the membranes, and perhaps also by quenching singlet oxygen (an excited sta te of molecular oxygen). The vitamin E radical formed is not very reactive.

Vitamin A (see review 38) and other retinoids also scavenge singlet oxygen- dependent reactions in membranes. There is also a growing interest in the utilization of these substances for the prevention of neoplastic transform ation of epithelial cells.

- 15-

1.6. Lipid peroxidation

Lipid peroxidation (8,32,39,59,77,152), is a phenomenon of central import­ance for the understanding of the damaging effects of free radicals. Biological m aterial contains a wide variety of polyunsaturated fatty acids (PUFA), in particular in membrane phospholipids. A characteristic property of such unsaturated fa tty acids is tha t in the presence of free radical initiators and oxygen they undergo oxidative deterioration. Food products as milk, cheese, and m eat fa t all contain large amounts of PUFA. When they undergo lipid peroxidation this is called rancidity and is accompined by a very unpleasant odour tha t makes food unpalatable.

A simplified scheme for the reactions involved in lipid peroxidation is shown in the figure in which R* is the radical initiator. In the first reaction,

R’abstracts hydrogen from the unsaturated fa tty acid (I) to yield the free radical (II). This is followed by the uptake of oxygen by (II) to produce the fa tty acid peroxy radical (III). This radical (III) reacts with another molecule of (I) in a propagation step, to yield a further molecule of (II) and the stable end product the unsaturated hydroperoxide (IV).

(-CH=CH-CH2-) + R- -------- (-CH=CH-CH-) + RH

(I) (II)

r-^(-CH =CH -CH -) + 0 2 (-CH=CH-CH-)I

(II) (III) O-O*

(-CH=CH-CH-) + (-CH=CH-CH2-)

O-O*(III) (I)

O-OH(IV) (II)

Figure A simplified scheme for the chain reaction involved in the formation of unsaturated lipid hydroperoxide (IV).

- 16-

Lipid peroxidation may be initiated by several types of free radicals among which are the oxygen free radicals. Lipid peroxidation is associated with a loss of unsaturated bonds, conjugation of double bonds (diene conjugation) and finally fragm entation with production of ethane, pentane and malon- aldehyde (152, 161). The biochemical consequences of lipid peroxidaton include perturbation of membrane architecture (8,152) with alteration of membrane function and permeability and enzyme inhibition. Examples of physiological consequences are hemolysis of erythrocytes, swelling of mitochondria, accumulation of lipofuscin in lysosomes, and generation of lipid peroxides in microsomes. It has been repeatedly shown that lipid peroxidation is much more likely to occur in membranes of animals who have been kept on diets deficient in vitamin E or selenium, and lipid peroxidation is also increased in animals fed a diet rich in polyunsaturated lipids (152,177). In vitro lipid peroxidation is greatly stim ulated by ferrous iron and reducing substances (8,39), and some chelating agents tha t bind iron may paradoxically increase lipid peroxidation (77).

1.7. Radiation and oxygen free radicals

In aqueous dilute systems such as the cell, most of the energy of radiation is deposited in water molecules, which lyse to yield water radicals. The primary radicals formed are the hydroxyl radical (OH*), the electron (e~), the hydrogen atom (H*), and some others in minor amounts (see table 1).

Table 1. Yields of the primary radicals and molecular products of water radiolysis a t neutral pH (46). (G = number of molecules transformed for each 100 eV of energy absorbed.)

Geäq+GH+ Geäq GH’ g OH‘ g H2 g h 2o 2

3.13-3 .31 2 .65-2 .83 0 .48-0 .64 2 .59-2 .75 0.45 0 .67-0 .72

When oxygen is present (as in oxygenated water) the hydrogen atom and the electron reacts rapidly with the oxygen molecules, Eq. 1-3.

- 17-

Eq. 1 H* + 0 2 ^ H02

Eq. 2 h o ;2 -►H+ + o :2

The reaction between H and 0 2, Eq. 1, generates perhydroxyl radicals (HCC), but a t physiologic pH this species rapidly loses the proton, and is transformed to the conjugate base 0 2, Eq. 2. Thus a t pH 7.4 the superoxide radical dominates over the perhydroxyl radical by a factor of almost 10^.

The hydrated electron directly forms superoxide with molecular oxygen,

The radicals formed through radiolysis may then react with solute mole­cules th a t are damaged, with more or less functional derangements of the cell and when radicals mediate the radiation damage, this is called the indirect action of radiation. Direct action is mediated, as the term implies, by the direct interaction between radiation and target molecules, e.g. DNA. It has been calculated in experiments using various radical scavengers tha t for cells such as the V 79 Chinese hamster fibroblasts, about 60% of the aerobic inactivation is related to OH* damage, 20% to d irect action and about 20% to other unknown mechanisms (33). In mouse L5178 Y cells about 70% of the single-strand breaks in DNA were related to hydroxyl radical or other indirect damage (146). Other values (both larger and smaller) for the relative contribution of direct and indirect radiation damage also exist using methods as freezing or drying thereby removing water radicals (5)

Since the 1950s it has been known that oxygen enhances the biological

effec ts of ionizing radiation by a factor of 2.5-3. Despite considerable research efforts the precise mechanism is still unknown. The most accepted theory, however, is the oxygen-fixation hypothesis (3), which sta tes tha t molecular oxygen by binding to the target (T) molecule transformed by radiation (T’), fixes the target molecule damage:

Eq. 3.

Eq. 3 e- + 0 2

- 18-

T + aerobic irrad ia tion---------- >*T' + 0 2 --------- ►TOO (schematically)

Under hypoxic conditions the transformed target molecule can be reconsti­tu ted by donation of a hydrogen atom from reduced glutathione.

T + hypoxic irrad ia tion ►T1 + GSH ►TH + GSSG (schematically)

Recently this theory received a strong support by the demonstration of a reduced ability of genetically glutathione deficient fibroblasts to withstand anoxic irradiation, and consequently to have a reduced OER (114).

1.8. Diethyldithiocarbamate (DDTC)

C H , C H j f^ N - C - S N a • 3H20

ch3ch2

Mol wt 171.21 (anhydrous); 225.26 (as trihydrate)

Diethyldithiocarbamic acid sodium salt: White crystals, m.p. 90-2°C; soluble in water, ethanol, methanol, and acetone, insoluble in benzene and ether. Relatively stable a t room tem perature in neutral or alkaline solution, but unstable in acid solution. Reacts slowly with atmospheric oxygen to form insoluble oxidation products. Chelates most divalent and trivalent cations (21). For a review on older literatu re on DDTC see reference 163.

Analysis of DDTC in biological m aterial has been performed using different analytical methods but today high performance liquid chromatography (HPLC) is probably the most sensitive and accurate method (89,121).

- 19-

1.8.1. Inhibition of CuZn superoxide dismutase by DDTC

In 1976 Heikkila e t al (83) demonstrated that DDTC could inhibit CuZn SOD both in vitro and in vivo, and la ter Heikkila e t al. (84) also described other compounds that were inhibitory. The DDTC inhibition was dem onstrated in vivo in the mouse by administration of 1.5g/kg body weight of DDTC intraperitoneally. By this procedure liver enzyme activity and blood enzyme activity were a t the lowest level (2 0 % of initial activity) one hour a fte r the DDTC administration and the enzyme activity then slowly returned. By 24 hours the enzyme activity in the liver was depressed to about 60% of initial value. In vitro the enzyme inhibition was shown to be essentially complete by incubation for 60 minutes with 2 mM DDTC (200 ng CuZn SOD/ml) and it could not be reversed by dialysis of the inhibited enzyme. However, enzyme activity was restored by dialysing in buffer

containing 0.5 mM CuSCty. Diethyldithiocarbamate formed a yellow complex with SOD, similar in appearance to the complex formed between DDTC and cupric ion complex. In contrast to inorganic Cu-DDTC, the complex between the Cu of CuZn SOD and DDTC was not extractable with organic solvents, which lead to the suggestion that DDTC was bound to the active site of the enzyme. Misra (118) suggested the formation of a ternary complex between enzyme-Cu and DDTC. Cocco e t al. (34), in a recent study repeated the experiments of Misra and reached a divergent conclusion, demonstrating that DDTC removed Cu from the active site of the enzyme, in a slow process, leaving the apoenzyme. The failure to remove the comlex by dialysis, gel filtration or solvent extraction was thought to depend on interaction of Cu-DDTC with the protein part of the enzyme, but not involving the native copper coordination ligands of the active site. Lengfelder (93) demonstrated tha t the DDTC-Cu complex has no superoxide dismutase activity.

1.8.2. Inhibition of enzymes other than CuZn SOD

Purified beef heart cytochrome oxidase was inhibited by 25% with 1 mM DDTC (72). A 50% inhibition of isolated ra t liver and beef heart cyto­

chrome oxidase activity by 5-1 OmM DDTC has also been reported (60).

- 20-

Metabolic activation of DDTC gives carbon disulphide, which can be oxidized by cytochrome P-450 to carbonylsulphide. This compound then releases atomic sulphur which can bind covalently to microsomal proteins in particular the cytochrome P-450 and destroy them. Therefore DDTC in vivo may inhibit cytochrome P-450. It has also been demonstrated that DDTC in vitro may inhibit benzo/a/- pyrene monoxygenase (BP monoxy- genase), and in vivo a reduction of both liver and intestinal BP monoxy­genase has been shown (67).

When DDTC is administered in vivo into rats the microsomal hydroxylation system is inhibited whilst the the glucuronic acid pathway enzymes are stim ulated (90,106).

It has been found that DDTC in vitro and in vivo inhibits dopamine-b-

hydroxylase and tyrosine hydroxylase (160).

Diethyldithiocarbamate can be enzymatically oxidized to its disuphide

form and subsequently inhibit many other enzymes, preferentially SH-group containing enzymes through formation of mixed disulphides, e.g. glucose hexokinase (156), and xanthine oxidase (56).

Inhibition of CuZn SOD leading to increased concentration of superoxide radicals has also been proposed to cause inhibition of glutathione per­oxidase in vivo and in vitro (64).

1.8.3. Effects of DDTC on subcellular organelles

U ltrastructural changes have been found in the mouse hepatocyte after DDTC feeding (1% in the diet) (7).Changes included reduction of liverweight and pale appearance of the livers. In the cells there was formationof megamitochondria evenly distributed in the hepatocyte, dilatation of the endoplasmic reticulum and disappearance of fa t and glycogen droplets. The mitochondria were too large to be accounted simply by a swelling of each mitochondrion. In the mitochondria the matrix and cristae were normal as far as their quantity and appearance were concerned. These authors found

- 21-

no other changes in cells apart from in the hepatocytes. In a study on malignant 3T3 cells similar changes in u ltrastructure were noted, however, a fte r incubation for 2 hours with O.lmM DDTC (96).

In another study DDTC (1.2g/kg) caused a marked reduction in ra t cardiac mitochondrial SOD, considerably decreased the oxygen consumption and respiratory control and induced a slower oxidative phosphorylation. Diethyldithiocarbamate also caused increased malondialdehyde content in mitochondria, indicating an increased lipid peroxidation of mitochondrial membrane (73).

Gallagher and Reeve (60) studying mitochondrial phospholipid synthesis noted tha t DDTC 3-1 OmM (as did copper deficiency) caused a reduction in

the binding of ADP to the inner mitochondrial membrane. This was proposed to impair the ATP production. The impaired ATP formation then would cause a decrease in the translocation of ATP to the outer mito­chondrial membrane, and a depression of ATP-dependent phospholipid synthesis.

It was demonstrated tha t DDTC induces synthesis of four proteins in normal chick embryo and human foreskin cells, and in the presence of copper DDTC was found to bind to HeLa cell DNA and Rous sarcoma virus 70S genome RNA (93).

1.8.4. Other biological effects

There are multiple studies of different biological e ffects of DDTC, some of these being initiated by the fac t tha t DDTC is the reduced monomer of Antabuse (Disulfiram). Mutagenic effects were studied in Ames test, and although ZnDDTC and CdDDTC showed direct mutagenic effects, pure DDTC was not tested, and disulfiram was not mutagenic (82). In several studies DDTC (and disulfiram) has been shown to have anticarcinogenic e ffec ts when administered before carcinogens as benzo/a/pyrene (67), and dimethylhydrazine (167). Diethyldithiocarbamate has also been used to prevent the hepatotoxic action of carbon tetrachloride (100,164). In rats

- 22-

DDTC prevented the nephrotoxic action of the cancer chem otherapeutic agent, cis-platinum (22) without abrogating its antineoplastic effec t (23). In ra ts and humans it reduces the toxicity of cadmium (37), nickel carbonyl (138) and mercury vapor (137). Humans exposed to nickel carbonyl vapours were cured without side effec ts with DDTC 1 g twice daily for 10 days (138), while a nontreated individual died of nickel poisoning. On the other hand DDTC has also been shown to have toxic effects on cells in culture (140) and on sperm atocytes (174) a t low concentrations. It has a peculiar biphasic toxic e ffec t on cells (41, 138). Various immunological effects of DDTC have been reported as the compound is related in structure to the thiol moiety of the immunoactive drug Levamisole (120). Diethyldithio- carbam ate is also used in neuropsychological and neuropharmacological research (63) as a noradrenaline depleting agent (due to the inhibitory action of DDTC on the enzyme dopamine-b-hydroxylase).

Diethyldithiocarbamate decreased (40,34) short term (48h) survival of rats in normobaric hyperoxic (93-97%) atmosphere, and of mice in ozone (64), which was related to the inhibitory e ffec t of DDTC on hyperoxia-induced increase in lung activities of SOD and glutathione peroxidase. However, DDTC was shown to increase long term (96h) survival of adult rats in 93- 97% oxygen and potentiated the inducing effects of oxygen on the

activities of glutathione peroxidase, glutathione reductase and glucose-6 - phosphate dehydrogenase without any effec t on the oxygen-induced increase of SOD (40).

Diethyldithiocarbamate is a radioprotector both in vitro and in vivo (4,12,17,170,171,). In these studies the compound was added just prior to the irradiation. The protective effec t was as good as that of cysteine or cysteamine (4). A fter the discovery of DDTC as an inhibitor of CuZn SOD, a radiosensitizing effec t of DDTC (93,96,143,133) was demonstrated in vitro, when the compound was added 30 to 90 minutes before irradiation. Diethyldithiocarbamate has also been found to protect against the lethal e ffec t of nitrogen mustard (29) in vivo, but in vitro it was shown to increase the toxic effec t of Bleomycin (97) and hyperthermia (93).

- 23-

2 PRESENT INVESTIGATION

2.1 Aims

The in terest in oxygen toxicity increased dramatically after the discovery of superoxide dismutase (SOD), as an obligatory defense against oxygen derived radicals. Oxygen also enhances the damaging effects of ionizing radiation and is therefore of great general in terest in basic radiobiological science and particularly in radiotherapy. In view of this, an experimental investigation on the possible significance of SOD and superoxide radicals in radiobiology and radiotherapy was of interest. In the present thesis, different experimental approaches were taken in investigating this concept.

A. To determine whether inhibition of intracellular CuZn SOD would increase radiation damage because of decreased protection against radia­tion-generated superoxide radicals. For this purpose cell culture using Chinese hamster V 79 fibroblasts (I,V) which contain CuZn SOD and Mn SOD, and human erythrocytes (IV) which contain only CuZn SOD were used.

B. To determ ine whether an increase in extracellular activity of CuZn SOD would decrease radiation damage because of increased protection against radiation-generated superoxide radicals. For this purpose, bovine CuZn SOD was added prior to irradiation (IV).

C. To assess the activities of CuZn SOD and Mn SOD in human tissues, tumours, and cell lines with the aim of detecting an obvious relationship between the enzyme activities and assumed (albeit not tested) radio­sensitivity.

- 24-

2.2 Materials and methods

A short description and some additional data on each method used in the publications (I-V) are presented in this section, but more detailed descrip­tions are found in the relevant publications or in the original publications on the methods given in the reference list.

2.2.1 Superoxide dismutase

Superoxide dismutase (SOD) activity (both CuZn SOD and Mn SOD) in homogenates (I-IV), was studied by a direct method (102), following the disappearance of superoxide radicals in a spectrophotom eter. The radicals were produced from small pieces of solid potassium superoxide a t pH 9.5, (the superoxide radical is more stable a t alkaline pH); and the enhancement

of the disproportionation reaction brought about by the addition of homogenates containing SOD was determined. The activity of Mn SOD was determined a fte r inhibition of CuZn SOD by 1 mM KCN, and CuZn SOD as to ta l SOD minus Mn SOD activity:

_ SOD2H+ + 2C>2 --------- >■ 0 2 + H20 2

This assay is direct and more sensitive than other assays, but is 10 times more sensitive for CuZn SOD than for Mn SOD. For this reason the Mn SOD activities in the thesis have been multiplied by a factor of 10.

Superoxide dismutase was also analysed by agarose gel electrophoresis in several of the tissues and the mesothelioma cell lines (III), and this sem iquantitative method agreed in every case with the enzymic activities.

The accuracy of the SOD assay was determined from a control sample consisting of the same human haemolysate (stored a t -80°C), over a period of almost a year. The standard deviation of the mean was about 5% both on a daily (n=54) and on a day-to-day (n=39) basis. This variation is valid for CuZn SOD and Mn SOD.

- 25-

2.2.2 Glutathione peroxidase

Glutathione peroxidase (GSHPX) in homogenates (III,IV) was determined according to Günzler e t al. (74), with some modificitions, by following the consumption of NADPH a t 340 nm in a spectrophotometer:

GSHPXROOH + 2GSH --------- ► ROH + GSSG + H20

GSSG is reduced back to GSH by the enzyme glutathione reductase (GSR):

GSRGSSG + NADPH + H+ --------- ► 2GSH + NADP+

The variation of this method was about 7% (SD) on a daily basis (n=10). In parallel determinations the variation according to Günzler (74) was about 3%.

2.2.3 Catalase

C atalase (CAT) was determined by two different methods. In (I) and (III) the catalase activity was measured by following the catalase-catalyzed initial liberation of oxygen from hydrogen peroxide, using an oxygen electrode, essentially as described by Del Rio (37). This assay is 50 times more sensitive than ordinary spectrophotom etric or titra tion methods. The variation of the method was around 10% on a daily basis (n=10).

CATH 2°2f H2 °2 ----------► ° 2 + 2H2 °

C atalase activity in the erythrocytes (IV) was determined by following the decrease of absorption of hydrogen peroxide a t 240 nm in a cuvette containing 10 mM hydrogen peroxide a t pH 7.4. The accuracy of this assay is sufficient for the determ ination of catalase in erythrocytes since the enzyme is abundant in these cells.

- 26-

2.2.4 Protein

Protein in the homogenates (I-IV) was determined using the blue colour compound Coomassie Brilliant Blue G-250 (27). This method is more sensitive, technically less complicated and more specific than the conven­tional Lowry method, and was shown to give comparable results in control experiments.

2.2.5 Cell culture

A substrain of the Chinese hamster cell line (V79-379A) was cultured in an ordinary CO2 cell culture incubator a t 37°C, pH about 7.4 and a t a relative humidity of more than 90%. The cells were grown in Eagle's minimal essential medium containing Earle's salts supplemented with 15% fetal calf serum and antibiotics in plastic (I) or glass (V) Petri dishes. For radiation

experiments, exponentially growing cells were trypsinized, monodispersed, counted and seeded into dishes of 6 cm diam eter. Treatm ent with DDTC was never started earlier than about 3 hours a fte r explantation, and irradiation was given about 5 hours a fte r explantation. Control dishes were trea ted in the same way. Control experiments with feeder cells demonstrated tha t the plating efficiency was not influenced by the number of cells initially explanted. Surviving clones were fixed, stained and counted a fte r 7-8 days. Colonies had to contain a t least 50 cells to be considered as representative of a surviving cell.

2.2.6 Cellular irradiation techniques

In experiments on clonogenic survival a fte r irradiation, the number of cells explanted were adjusted according to each radiation dose in such a way th a t 50-100 cells would be expected to survive. In no case were more than 15.000 cells allocated to a 6 cm dish. Before the irradiation the cells in (I) were gassed with CO2 a t 37°C, in tight plastic boxes, and then transferred

to the radiation departm ent for irradiation. The tem perature during the irradiation was around 30°C. In the so-called postirradiation experiments

(V), the cells were irradiated a t room tem perature without CO2 gassing. In experiments determining clonogenic survival under various conditions of oxygenation (V), the cells adhering to P etri dishes in only 1 ml of medium

- 27-

were irradiated a t room tem perature; the small quantity of medium facilitating gas exchange. The dishes with their covers removed were placed in an air-tight plastic box and exposed before and during irradiation to a gust of either argon, containing less than 2 ppm oxygen impurity (total exposure time 15 min.), air (5 min.) or pure oxygen (5 min.). Carbon dioxide was added to the gas flow to maintain pH a t a normal level.

2.2.7 Erythrocytes

Erythrocytes (IV) were collected from myself, washed three times with Eagle's MEM and diluted to a haem atocrite value of 1/20. This suspension was then treated with DDTC as described below. In irradiaton experiments aliquots of 0.4 ml were added into wells (diameter 2cm) in a multiwell dish, irradiated and the cells incubated overnight in a refrigerator before the

determ ination of radiation-induced haemolysis.

2.2.8 Radiation-induced haemolysis

Erythrocyte haemolysis (IV) was determined by the median glycerol haemo­lysis procedure of Gottfried and Robertson (66). Irradiated erythrocyte suspension (20-50 ul) was added to 3 ml 0.2M glycerol in 50 mM NaCl buffered with 3 mM potassium phosphate, pH 7A . The haemolysis was monitored as the fall in turbidity a t 625. The median glycerol haemolysis tim e (GLT 50) was taken as the tim e required for the optical density to fall to half the initial value.

The mean GLT 50-value of control erythrocytes when assayed immediately a fte r sham-irradiation was 53.5 ± 1.5s (SD, n=6)on a day to day basis. The GLT 50-value of control cells a fte r sham -treatm ent and 20 hour incubation in the refrigerator varied more, 44.3 ± 3.7s (SD, n=19) between the experiments. The mean difference between double determinations of GLT 50 varying from 16.8 to 47.0 s within the radiation experiments, was around 2 per cent.

- 28-

2.2.9 Radiation dose-effect curves

The survival curves from the experiments on the V 79 cells (I,V) were plotted for each experiment. In these curves the logarithm of the surviving fraction is represented on the y-axis and radiation dose is represented on the x-axis. The survival curve may also be represented by an equation, and many different equations exist, baséd on different theories and yielding different param eters for the equations. The most commonly utilized equation, based on the so-called single-hit m ultitarget theory, is used in the present investigation:

S = 1 - (1 - e-D/Do)n

In this equation S is the surviving fraction, Do is the radiation dose tha t causes a decrease of the surviving fraction by a factor 1/e (about 37%) along the final straight portion of the curve. D is the actual dose administered and n the extrapolation number, is defined as the in tercept of the extrapolated straight portion of the curve on the y-axis. The dose- modifying factor (DMF) or dose-modifying effect, is often defined as the ratio between the slope constants or Do-values of the straight parts of two survival curves in paired experiments. When oxygen is used as a modifying factor the DMF of oxygen is called the oxygen enhancement ratio, OER.

In erythrocyte experiments the logarithm of the GLT 50 (as percent of initial GLT 50) was plotted against radiation dose and the slope and intercept values were calculated by linear regression analysis.

2.2.10 Radiation split-dose survival

Split-dose experiments were used to study the capacity of the irradiated cells to recover from so-called sublethal radiation damage (48). In such experiments the surviving fraction a fte r a single dose treatm ent is compared to the surviving fraction after administering the same dose, dividid into two equal fractions (i.e. half the single dose administered twice). As expected, the survival after split-dose treatm ent is higher than a fte r single dose treatm ent. The enhancement of survival is usually

- 29 -

presented as the split-dose ratio, which is defined as the surviving fraction a fte r split-dose treatm ent divided by the surviving fraction after single­dose treatm ent, for example:

Split dose ratio = Survival a fte r 2 x 4.0 Gy/ Survival a fte r 8.0 Gy

By varying the tim e-interval between the split-dose fractions the kinetics of recovery can also be studied. In the split-dose experiments in (V) the tim e-interval between the split-doses was 18 hours. This interval has previously been found to be shorter than the tim e necessary to complete a cell division under the prevailing conditions (98).

2.2.11 Statistics

Statistical testing of the differences between the slopes and intercepts were performed by (one-way) analysis of variance in (I), by multiple regression analysis in (IV) and by Student’s t- te s t in (V). The results of (I)

has also been evaluated la ter by Student's t- te s t and multiple regression analysis with no change of the conclusions made in (I).

2.2.12 Diethyldithiocarbamate (DDTC)

The DDTC solution was freshly prepared every day. Since the pH of the solution was about 7.8, no relevant decomposition of DDTC was possible for several hours (21).

2.2.13 Diethyldithiocarbamate-inhibition of superoxide dismutase

Essentially the same procedure was used when treating V 79 cells and erythrocytes with DDTC. The ordinary medium in which cells were suspended was exchanged with medium containing 3mM DDTC, followed by incubation for 90 min in an incubator. In the erythrocyte experiments DDTC was added to the erythrocytes to give a final concentration of 3mM. DDTC was then removed by 4 successive medium changes (in erythrocyte

experiments centrifugations were used), w hereafter the cells were irradiated. In the experiments on postirradiation treatm ent of V 79 cells the irradiation was given a t various intervals before the DDTC-treatm ent (V).

- 30-

For the determination of SOD activity in DDTC treated V 79 cells, several cell culture dishes were treated with DDTC and a t appropriate time intervals the cells were washed twice with buffered saline containing a diethylenetriam ine pentaacetic acid (DTPA), a copper-chelating substance (to prevent enzyme reactivation), scraped off and stored a t -80°C until enzyme assay. A fter 90 min DDTC was removed as described above. Control plates were treated identically. In the case of erythrocytes it was sufficient to use only one dish and take aliquotes (0.4ml) a t predetermined intervals during and a fte r the DDTC treatm ent.

2.2.14 Removal of DDTC

Removal of DDTC appeared essential in the radiation experiments as free DDTC in vivo and in vitro is an efficient radioprotector, and as removal would minimize effects on other enzymes in the cells. The free DDTC concentration a fte r the washings therefore was determined in separate experiments (I), where about 15x10^ cells were exposed to 15 mM DDTC for 90 minutes. The remaining DDTC afte r successive medium changes was estim ated by spectrophotom etry from homogenates, a fte r extraction with isoamylalcohol. A fter four washings no remaining free DDTC (the lower detection level of the assay was 50 uM) could be found in the cell homogenates. In the radiation experiments, as a few thousand cells were exposed to only 3 mM DDTC, the free DDTC concentration in these cells would have been even lower.

2.2.15 Cell cycle analysis of DDTC-treated cells

In order to assess other effects of DDTC on the cells of possible importance for radiation response, separate experiments were performed aiming to detec t a cell-cycle effec t of DDTC on the V 79 cells. Such an e ffec t of DDTC could change the radiosensitivity (76) of the cells.

In these experiments about 1 million cells were seeded into the Petri dishes, incubated for adherence for 3-4 hours, and then incubated with 3 mM DDTC for a further 90 minute period. After removal of DDTC by four successive medium changes, the cells were washed twice with phosphate

- 31-

buff ered saline and 1-2 ml of a propidium iodide solution containing RNAse and a detergent (173) was added. DNA analysis was performed by flow cytofluorometry as described by Roos e t al. (145). The fraction of cells in the different phases (G0-G1,G2-M and S) of the cell-cycle and the fraction of diploid cells were determined from the DNA histograms of DDTC- treated and untreated cells. The results from 4 separate experiments showed no significant difference between DDTC-treated and untreated cells. See table 2.

Table 2. DDTC effects on the distribution of V 79 cells in the various phases of the cell cycle and the fraction of diploid cells. Figures are mean and SEM of four paired experiments.

DDTC O mM DDTC 3 mM DDTC 0 mM DDTC 3 mMGq-G i S g 2- m Gq-G i S G2-M Diploid cells Diploid cells

% % % % % % % %

Mean 64.0 24.0 11.9 62.0 23.5 14.5 89 90

SEM 1.5 2.1 2.6 0.9 1.6 2.4 1.5 1.8

2.2.16 U ltrastructural effects of DDTC

The effec t of DDTC on ultrastructure of V 79 cells was investigated in a separate experiment. In this experiment about 5 million cells were used and treated with DDTC for 90 minutes as described earlier. A fter removing the DDTC and washing the cells twice with isotonic NaCl the cells were fixed in 2.5% glutar aldehyde in 0.1 M sodium-cacodylate + 0.1 M sucrose, a f te r­fixated in 1% osmium tetroxide in 0.15 M sodium-cacodylate, embedded in Epon, sectioned and finally contrasted with uranyl ace ta te and lead c itra te . A JEOL 100 CX electron microscope was utilized for morphological assessment.

No ultra-structural aberration could be detected in the cells treated with 3 mM DDTC compared to the DDTC-untreated cells, see figure 5.

- 32-

Figure 5a. Electron micrograph of a DDTC-untreated Chinese hamster V 79 cell. M = mitochondrion, N = nucleus.

Figure 5b. Electron micrograph of a Chinese hamster V 79 cell trea ted with3 mM DDTC for 90 min.

- 34-

2.2.17 Cell lines

Cell lines used in (III) consisted of 31 human cell lines of different origins. Most lines are well known and are characterized in a number of earlier investigations. The mesothelioma cell-lines were established from pleural effusions and have neoplastic properties, as judged by chromosome analysis and cell growth characteristics (154). The 3C-1 line, derived from a giant cell tumour, had reached passage 21 and could not yet be considered as a permanent cell line. The cells were grown in Eagle's minimal essential medium (anchorage dependent lines) or Ham's F-10 medium (suspension growing lines) supplemented with 10% fe ta l calf serum, and with anti­biotics. The cells were harvested 3-7 days a fte r explantation, washed twice with 0.15 M NaCl containing 3 mM DTP A, centrifuged and stored a t -80°C until assay of enzymes.

2.2.18 Tissue specimens

Tissue specimens were collected from accident victims without known physical disease within 24 hours afte r death. Homogeneous tumour pieces were cut out of surgical specimens within 1 hour after their removal from the patients. The tissues were then stored a t -80°C until homogenized.

2.2.19 Correction for tissue erythrocyte contamination

Erythrocytes contain large amounts of three of the enzymes assayed in (III). The content of glutathione peroxidase and especially catalase is larger than in several of the tissues. It was, therefore, necessary to compensate for the enzyme activities in the homogenates contributed by the contami­nating erythrocytes. The hemoglobin was specifically determined (103) in term s of the differential protective activity of serum (haptoglobin) and egg white on the peroxidatic activity of the hemoprotein. Correction of the erythrocyte contamination of tissue homogenates was thus made according to the enzymic activities of the erythrocytes from the corresponding individuals.

- 35-

2.2.20 Post-mortem stability of tissue enzymes

The possibility of a decrease of enzyme activities in tissues before freezing was investigated in experiments on swine liver. Small pieces of the liver were removed from an anesthetized animal, and kept in te s t tubes (in the dark) a t room tem perature (+22°C) and in a refrigerator (+4°C) for various periods of tim e, a fte r which the tubes were transferred to -800C. Enzyme activities were determined la ter on the thawed samples a fte r homogeniza­tion and sonication according to the procedures utilized in (II) and (III). No enzyme inactivation was seen for a t least 30 hours a t either tem perature, see table 3. Similar findings have been published in a study on the post­mortem stability of ra t brain catalase and glutathioneperoxidase (28).

Table 3. Post-mortem stability of swine liver CuZn superoxide dismutase, Mn superoxide dismutase, catalase and glutathione peroxidase a t two different tem peratures. Each figure represents the mean of two samples from the same liver.

Enzyme activ ities (10^ x units/mg protein)Time a fte r CuZn SOD Mn SOD Catalase GSH-peroxidasedeath (h) +4°C +22 +4 +22 +4 +22 +4 +22

0 87.90 87.90 1.35 1.34 155.40 155.40 13.20 13.20

6 93.40 97.70 1.40 1.53 158.10 167.60 15.90 14.10

18 97.00 90.60 1.44 1.46 167.40 155.90 15.60 11.70

30 84.30 104.10 1.46 1.42 163.10 153.80 14.50 12.80

2.2.21 Strength of extraction medium and enzyme activities

The variation of enzyme activities due to different strengths of the extraction media was studied on mouse kidney homogenates. The strength

of the extraction medium was increased by addition of extra KC1 or by addition of a detergent (Triton X-100). The results are presented in table 4. Only in the case of glutathione peroxidase an increase of the strength of the extraction medium enhanced the yield of the enzyme.

- 36-

Table 4. Influence of extraction medium strength on the yield of enzyme activity from mouse kidney homogenates. Each figure represents the mean of two samples.

Extraction medium

Centrifuga-tion

- Enzyme activities (10^ units/g wet weight)

CuZn SODMn SODCatalase GSH-peroxidase

40.7 2.27 33.5 90.5I 10 mM K-phosphate 37.2 2.56 36.6 89.1

pH 8.0 + 30 mM KC1 36.7 2.52 35.4 71.6+ 42.0 2.58 36.7 74.3

38.6 2.06 37.3 90.5II Same as I, plus 37.2 2.41 39.1 90.5

extra 250 mM KC1 38.8 2.60 37.1 87.8+ 39.3 2.60 39.9 86.4

43.0 2.60 39.4 87.8III Same as I, plus 42.5 2.54 37.8 -

0.3 % Triton X-100 39.3 2.33 35.4 82.4+ 40.4 2.81 35.1 82.4

2.2.22 Strength of sonication and enzyme activities

To test the influence of variations in the homogenization and sonication procedures on enzyme activities, two separate experiments were done, using the suspension growing cell-line Namalwa (1-7x10? cells), that had been rinsed four times with 0.15 M NaCl containing 3 mM DTP A, collected and stored a t -80°C. For sonication, the cells were then thawed and mixed with 1.5-4 ml, 10 mM K-phosphate buffer pH 7.4 + 30 mM KC1 in test tubes and sonicated on ice for one minute. A Branson Model B 30 cell disruptor, operated with the micro tip a t 10% 'duty cycle', with 'output control' varied from 1.4-4, and a t 'continuous' mode was used for sonica­tion. A fter sonication the cell suspension was left in air a t +8°C for 30 minutes for extraction of enzymes, centrifuged, and enzymes in the supernatant assayed, see table 5.

- 37-

Table 5. Influence of differences in sonication conditions on the activities of superoxide dismutases, catalase and glutathione peroxidase in homo- genates of Namalwa cells. Figures are mean and SEM of 13 runs (experi­ment No. I) and 5 runs (No. II).

CuZn SOD Mn SOD C atalase GSH-peroxidase Experiment No. (U/mg protein) (U/mg protein) (U/mg protein) (U/mg protein)

I Mean 83.6 1.87 45.7 50.4SEM 5.8 0.23 8.53 9.50

II Mean 95.7 1.96 33.9 57.9SEM 13.6 0.10 4.40 5.60

As can be seen in the table the difference between the means of the first and second runs of the experiment were for CuZn SOD +14%, for Mn SOD

+5%, for catalase -26% and for glutatione peroxidase +15%.

In conclusion the inherent variation in the determinations of the enzymes seem to be about 5-8%. The variations when the to tal procedure is included is much larger and for catalase and glutathione peroxidase may amount to 25%. However even if this variation seems large, it cannot explain the large variations in enzymic activities in the tissues, tumours and cells.

2.2.23 Irradiation

In (I) and (V) irradiation was produced from a 4 MV linear accelerator (Varian Clinac 4) with a dose ra te of 3 Gy/min. In (IV) irradiation was produced from a 50 kV X-ray machine with a filtration of 1.0 mm Al, HVL 0.9 mm Al and a dose ra te of 12 Gy/min a t the surface of the solution. In (V) irradiation was produced from a 190 kV X-ray machine with 1.5 mm Al inherent filter, 1.0 mm Al additional filter, HVL 0.95 mm Cu, a dose rate of 235 R/min and a focus to ta rget distance of 43 cm.

- 38-

2.3 Results and discussion

2.3.1 Radiation experiments

Superoxide radicals are generated in cells through autooxidative and metabolic mechanisms. Radiolysis of water also yields a large amount of superoxide radicals. Superoxide radicals as well as radiation have been shown to damage DNA, virus, bacteria and higher cells.

Superoxide dismutase is the principal defense in the cell against the superoxide radical and it was therefore of in terest to investigate the possible significance of an inhibition of CuZn SOD for the damaging effects of radiation.

Inhibition of CuZn SOD and specificity of the inhibition

It was demonstrated (I,IV) tha t in both fibroblasts and erythrocytes CuZn SOD was inhibited by more than 95% by treatm en t with 3mM DDTC for 90 minutes. Furthermore enzyme inhibition in both cell types was not reversed a fte r removal of excess DDTC. At the s ta rt of irradiation, about 30-45 minutes a fte r the end of the DDTC exposure, CuZn SOD activity in the cells was reduced by more than 90%; and in the following two hours the enzyme was still more than 80% inhibited.

Since DDTC has a wide range of biological effecs (see section 1.8.), other

factors of importance for the radiosensitivity may have been influenced. On the other hand, if DDTC was found not to influence such factors the

relevance of the CuZn SOD inhibition for the radiation effects would increase.

Thus Mn SOD (I), catalase (IV, 104), glutathione peroxidase (V) and cyto­chrome oxidase (104) activities were not changed by the present DDTC procedure and neither was the content of non-protein sulphydryl (104). There was no change in cell morphology as seen by electron microscopy (see section 2.2.16.), and no influence on distribution of cells in different phases of the cell-cycle as determined by flow cyto-fluorom etry (see section 2.2.15.).

- 39-

Radiation clonogenic survival a fte r CuZn SOD inhibition

P retreatm ent with DDTC decreased the clonogenic survival of V 79 fibro­blasts (I) as seen by a reduction of the Do value from 2.29 Gy in control cells to 1.87 Gy in DDTC pre-treated cells giving a DMF of 1.23. There was no influence on the extra-polation number. The results, therefore, were compatible with the proposition tha t superoxide radicals contributed to radiation damage.

Workers using Chinese hamster (DON) cells and mouse 3T3 cells (95,96), and using lymphocytes (143) have reported a radiosensitizing e ffec t of DDTC. However, in contrast to the present investigation, they did not remove DDTC before irradiation and neither was the actual magnitude of SOD inhibition determined.

Significance of Mn SOD?

The Mn SOD activity in the V 79 fibroblasts contributed to only 20% of the to ta l SOD activity before DDTC treatm ent. In contrast, a fte r DDTC trea tm en t which had no e ffec t on the activity of Mn SOD, this enzyme accounted for the largest part of functional SOD. Most of the Mn SOD is localized in the inner membrane and matrix space in the mitochondria, but in prim ates Mn SOD is also found in the cytoplasm. In ra t liver cells about 20% has been found in the nuclear fraction (168). The distribution of Mn SOD in the V 79 cells is not known, and it cannot be excluded tha t Mn SOD is also present in in the cytoplasm and the nucleus of these cells.The superoxide anion is comparatively stable and as it perm eates cell membranes it may be scavenged by Mn SOD not only in the mitochondria but also in other cell com partm ents. It is, therefore possible that though most of the CuZn SOD in the cells was inhibited by DDTC, a more complete SOD inhibition (including also inhibition of Mn SOD), could have brought about an even larger enhancement of radiation damage.

Trypsinization has been found to slowly reduce Mn SOD activity in normal WI-38 cells (human fibroblasts), but not in the SV-40 transformed WI-38 cells (177). In V 79 cells, which are transformed cells, trypsinization would

-4 0 -

probably exert the same effects as in transformed WI-38 cells. On the basis of this, this method of reduction of Mn SOD could not be utilized. The investigation on cell lines (III), showed that Mn SOD was not reduced in any of the cell lines either. As erythrocytes are known to lack Mn SOD, experiments were carried out to determine whether DDTC treatm ent would induce an even larger enhancement of radiation damage in these cells.

Radiation response after CuZn SOD inhibition of Mn SOD deficient cells

Since clonogenic survival could not be used as an endpoint in erythrocytes, haemolysis was chosen as the indicator of radiation damage (see IV). Pre­treatm ent with DDTC of erythrocytes caused a substantial reduction in their CuZn SOD activity, now representative of the reduction of to tal SOD

activity. Despite this, only a small increase in radiation damage was noted. It appeared as an increase of the slope of the dose response curve of the DDTC pre-treated cells giving a DMF of 1.04. (not significant), and a reduction of the intercept from 98.6% (of initial GLT 50) for control cells to 92% for DDTC pre-treated cells (p=0.005). These results were in contrast to the DDTC-enhancement of radiation haemolysis observed by Stone e t al. (155).

Increasing the extra-cellular CuZn SOD activity before irradiation

Experiments aimed a t determining whether addition of exogenous enzymes SOD and catalase would increase the radiation resistance of erythrocytes were carried out. It was also hoped that the addition of these enzymes would counteract a possible radiosensitizing e ffec t of DDTC pre­treatm ent. This was, however, not found as only a small and insignificant reduction of haemolysis of both control, and DDTC pre-treated cells was observed. The results were thus very similar to those from the DDTC pre­treatm ent experiments and indicated tha t superoxide radicals are of only marginal importance for radiation haemolysis. Other studies (16) are in agreem ent with this, emphasizing instead the significance of the hydroxyl

radical. Furtherm ore, the assumption that erythrocyte deficiency of Mn SOD should further increase the enhancement of radiation damage after DDTC pre-treatm ent was found to be wrong.

- 41-

Importance of CuZn SOD for the radiation response of different cell types

The erythrocyte is a highly specialized nonproliferating, anucleate cell. In view of this, the effec ts of DDTC on cell com partm ents and cell functions of nucleated cells, such as the V 79 cells are not representative for specialized cells like erythrocytes. Moreover, radiation haemolysis is determined within 24 hours in a metabolically inactive cell, whilst clono- genic survival is determined afte r 7 days in proliferating, metabolically active cells. A decreased defence against superoxide radicals was, there­fore, still a possible explantion for the enhanced aerobic radiation damage in DDTC pre-treated V 79 cells. However, participation of superoxide radicals in the immediate radical-dependent, physico-chemical radiation damage has been questioned by fast-kinetic studies (112), and by studies on T4 bacteriophage (149) and E. coli B (150). A ttem pts to reduce radiation

damage in vivo and in vitro in mice or mouse cells respectively (131), by increasing the amount of extracellular SOD before irradiation have also failed. Even so, DDTC pre-irradiation treatm ent may be radiosensitizing, since the intracellular CuZn SOD activity, and thereby the defense against superoxide radicals was also markedly reduced during the first hours of the post-irradiation period.

Protection by SOD against immediate or delayed radiation damage?

Since the presence of oxygen during water radiolysis results in the formation of superoxide radicals from the hydrated electrons and the hydrogen atoms (see section 1.7), radiolysis in the absence of oxygen cannot yield superoxide radicals. To study the importance of superoxide radicals in the immediate e ffec t of radiation, an a ttem pt was made to prevent DDTC-enhancement, by anoxic irradiation of DDTC pre-treated cells. Irradiation in oxygen was also tried in order to further enhance the e ffec t of DDTC.

In these experiments (V), however, DDTC-enhancement in argon, air or oxygen did not differ significantly, with a DMF in argon of 1.31, in air of 1.26 and in oxygen of 1.51. Accordingly, DDTC-enhancement could not be dependent on the generation of superoxide radicals during the irradiation

- 42 -

period suggesting that superoxide radicals do not seem to be associated with immediate radiation damage. The e ffec t in argon was still compatible however, with increased damage from superoxide radicals, as in these experiments the cells were returned to oxygenated atmosperic conditions within 30 seconds after the end of the irradiation while CuZn SOD continued to be markedly inhibited for several hours a fte r the irradiation.

Protection by SOD against delayed radiation damage

The results with the cells irradiated in anoxic conditions suggested that DDTC-enhancement was a post-irradiation enhancement. This in principle could be caused either by DDTC-interference with sublethal type of radiation damage (48), or by interference with so-called potential lethal damage (PLD), (138).

The results from experiments showing survival curves for DDTC pre­treated and untreated cells in air (I,V), argon (V) and oxygen (V), which had shown no effec t of DDTC pre-treatm ent on the extrapolation numbers, however, indicated that DDTC did not in terfere with the expression/repair of sublethal radiation damage. This was in accordance with the results of the split-dose experiments in (V), since the increase of the split-dose ratio of cells trea ted with DDTC only before the first fraction was incompatible with a DDTC-effect on sublethal radiation damage (see discussion in V).

Post-irradiation treatm ent of the cells with DDTC was therefore tried a t

various intervals a fte r irradiation, and a post-irradiation enhancement was detected. The enhancement of radiation damage was of the same magni­tude as could be seen from survival curves after DDTC pre-treatm ent, with a reduction of survival by a factor of about 2 a t the dose level of 7.0 Gy. The e ffec t was seen to diminish with tim e and had completely disappeared within 14 hours, probably even sooner.

The enhancement induced by DDTC is, therefore, a post-irradiation phenomenon, related to potentially lethal radiation damage, which can be accomplished either by treatm ent of the cells with DDTC immediately

- 43-

before the exposure to radiation, or by DDTC-treatm ent in the post­irradiation period. The e ffec t is seen as an increase of the slope of the survival curve by a DMF of about 1.3 whilst there is no change in the extrapolation number.

Further indications of a post-irradiation importance of SOD

Addition of SOD protects isolated bacteriophage DNA from inactivation by radiation (172), and increases viability and preserves morphology of irradiated foetal calf myoblasts (113). However, in these studies form ate was added before irradiation to convert all radicals to superoxide radicals.

It has been reported (134) that addition of SOD to the medium reduced radiation damage to mycoplasma, and the maximal protective e ffec t was seen when the enzyme was added 10 min afte r irradiaton. Phospholipid membranes produced from soybean lipid fraction are also found to be protected by SOD, principally when this is added before irradiation, but some protective effec t is also seen when SOD is added 5 minutes a fte r irradiation (136).

Several studies, not always differentiating between immediate and delayed

effects, also indicate the existance of toxic superoxide-dependent processes afte r irradiation. A reduction in the oxygen enhancement ratio (OER), a t least when a dilute suspension of stationary-phase E. coli B is irradiated in the presence of SOD has been reported (117). The OER with SOD decreased from 2.3 to 1.3, but boiled, inactivated enzyme had no e ffec t. Catalase, mannitol and histidine, so-called hydroxyl radical scavengers, were as effective as SOD. Other workers found an increase in OER in E.coli B cells containing lowered activities (about 60% of normal)

of periplasmic Fe SOD of this bacteria, and this increase in OER could be prevented by addition of exogenous SOD to the medium before the irradiation (124).

There are also reports of SOD, catalase, or both, diminishing the number of radiation-induced chromosome abberations in human lymphocytes; not

- 44-

depending on lymphocyte cell-cycle phase and being even more pronounced when the enzymes were added afte r the irradiation than before the irradiation (122,123).

Neoplastic transform ation of mouse embryo fibroblasts after irradiation or treatm ent with Misonidazole, have been found to be reduced if SOD is present during the post-irradiation/post-treatm ent period, but not if SOD is present only for a short period before or during irradiation/treatm ent with Misonidazole (113).

The results of Petkau e t al. (133,137), show that SOD can raise the LD 30/30 of six-weeks old female Swiss mice by a factor of 1.33, and the most effective treatm ent is to inject the enzyme both before and a fte r the

irradiation. In agreem ent with this Abe e t al., using a different mouse strain, found an increase of LD 30/30 but only with a DMF of 1.03. (1). Recently, Gray and Stull (69) described a protection by polyethylene glycol-attached CuZn SOD (PEG SOD) and PEG-catalase on mouse LD 30/30, but the e ffec t was dependent on the mouse strain employed.

A transferable chromosome-damaging e ffec t of plasma obtained from accidentally or therapeutically irradiated humans has been detected when this plasma is added to normal lymphocytes (63,86,131). This has been shown by Emerit (49) and in several studies she has shown that blood taken from patients suffering from auto-immune diseases, inflammatory bowel

diseases, and congenital chromosome breakage syndromes, e.g. Bloom's syndrome and ataxia telangiectasia, also have a transferable clastogenic (chromosome breaking) e ffec t. The effec t seems to depend on the presence of a low molecular weight substance generated in the blood when it is exposed to superoxide radicals. The clastogenic effec t on normal cells is, therefore, reduced in the presence of SOD (49).

Accordingly, these studies using addition of exogenous SOD to reduce radiation damage in the post-irradiation period in different systems, are in agreem ent with the results of the present investigation where radiation

- 45-

damage is enhanced by reduction of intracellular CuZn SOD activity during

the post-irradiation period.

Conclusions

A procedure for inhibition of intra-cellular CuZn SOD by DDTC was developed. An increased radiation damage of metabolically active cells was seen a fte r such a treatm ent. The increased damage is compatible with the concept of an increased superoxide-radical mediated damage in the first post-irradiation hours. The post-irradiation enhancement is also in agreem ent with investigations suggesting tha t superoxide radicals do not participate in the immediate damaging effects of radiation (112,149,150).

As to the mechanism of the post-irradiation enhancement one can only speculate. It could involve increased metabolic production of superoxide radicals and derived species in irradiated cells compared to unirradiated cells, membrane lipid peroxidation, and extra-nuclear and nuclear events.

2.3.2 Superoxide dismutase in tissues, tumours and cell-lines

Many studies have been published on the activities of SOD and the other protective enzymes in tumours and tumour cells because of the oxygen radical scavenging abilities of these enzymes and the possibility of a relationship between SOD and radiosensitivity. The present study examined the activities of CuZn SOD and Mn SOD in human tumours and tumour cells, and for comparison in normal human tissues and cells. Catalase and glutahione peroxidase, which also are of conceivable in terest in connection with free radicals were determined in the tissues and the cell-lines, but not in the tumours. The present study is much more extensive than most other studies published which have been performed only on single tumour specimens, or on experimental murine tumours.

Activities of CuZn superoxide dismutase

Tumours in general (II), had CuZn SOD activities around 10.000 U/g wet weight, which was low compared to metabolically more active normal tissues as the liver (100.000 U/g) and kidney (30 - 40.000 U/g), but similar

-4 6-

to most other organs and higher than adipose tissue (1.000 U/g). The high enzyme activity of the liver should indicate a very high radioresistance, but this is controversial, some studies have reported the liver to be only moderately radioresistant (88,90). On the other hand, skeletal muscle with a medium CuZn SOD activity of 11.000 also has been considered "radio­resistant".

Most of the tumours had enzyme activities around 10.000 and there were several tumours, e.g. the lymphomas, and the Ewing sarcoma, tumours which usually are rather radiosensitive, tha t had quite similar enzyme levels to tumours which are generally more radioresistant like the malignant melanoma.

This suggests tha t there is no direct correlation betweeen tumour or tissue

enzyme activities and radiosensitivity. However the mechanims that lead to destruction of tissues and tumours afte r irradiation are complex depending also for example, on the proliferative sta te of the cells and the vascularisation. Tissues are also a mixture of different cell types, with different enzyme activities and possibly different radiosensitivities. It was therefore of considerable in terest to study the enzyme activities in more pure systems such as cell cultures.

This investigation approach (III) demonstrated that CuZn SOD activities did

not on the whole vary much between the different cell lines, which was in accordance with the suggested obligatory requirement for the enzyme in aerobic cells. The to ta l mean for the enzyme activity was 178 U/mg protein with a SEM of 11 U (6%). The highest enzyme activities were found in the Jurkat, and the RPMI 8226. lines with CuZn SOD activities of 300 U, and the lowest activities were found in the embryonal lung fibroblast with 83 U, and the renal carcinoma cell line HCV 29 with 72 U/mg protein.

Thus the findings of reduced activities of CuZn SOD (cytosolic SOD) as reported in studies on transplantable murine tumours (15,20,25,126,148), and in a study on a human tumour (58), was not substantiated here. Recently increased activities of CuZn SOD in tumours and tumour cells

- 47-

have also been reported (178, 99). Furthermore, no relationshipe was found between the enzyme activities and assumed radiosensitivity in the present studies.

A ctivities of Mn superoxide dismutase

The Mn SOD activities in tissues and tumours were much more variable than the activities of CuZn SOD. Mn SOD was present in all of the tumours investigated, most of the tumours containing from 1.000 to 10.000 U/g wet weight (determined Mn SOD activities multiplied by 10, see section 2.2.1.), but some of the sarcomas contained less than 1.000 units. These enzyme activities were within the range of several normal tissues, where again the liver contained the highest enzyme activities (around 25.000). Erythrocytes were shown to lack Mn SOD, and adipose tissue contained rather little enzyme.

The investigation on cell lines yielded a similar picture with a large Mn

SOD variation and with exceptionally high activities in two of the mesotheliomas, P27 with 2300 units/mg protein and P7 with 1260 U. The

median enzyme activity was tha t of the glioma cell line 251 MG, with an enzyme activity of 57 U and the lowest activity was tha t of the T- lymphoblastic cell line Molt 4 with 13 U. The mean value of all of the 31 cell lines was 258 U with a SEM of 84 U (30%). The variability of the Mn SOD activities in cells and tissues may be related to the known inducibility of this enzyme (47,70,110).

There was no correlation between Mn SOD activities and CuZn SOD activities and again no direct relationship was found between the Mn SOD activities and the assumed radiosensitivities of the tissues, tumours or cell lines. A more reliable way of investigating the proposition would, however, be to determine the radiation sensitivity of the cell lines in parallel to enzyme determinations. Such studies are a t present in progress.

Reduced or a to tal lack of Mn SOD (mitochondrial SOD) activity has been reported on repeatedly, mostly in transplantable tumours (20, 42, 99, 126,

- 48-

127, 148, 169) but also in human cells in culture (177) and in human leukemia cells in vivo (178). In two recent publications (15,99), a direct correlation between the growth rate of experimental rodent tumours and the decrease of Mn SOD activity was found. Even a proposition (125,128), that all neoplastic cells lack Mn SOD, and that this lack of Mn SOD is a prerequisite for malignant transform ation, has been put forward. The present investigations, which were performed with human m aterial do not support this proposition. Other studies have also found Mn SOD activities in tumours and tumour cells (99,127,178).

Total superoxide dismutase activities

Several studies have only measured to tal SOD activity. Reduced total activity of SOD has been reported in several transplantable mouse, ra t and

hamster tumours (13,20,30,127,169), in transformed cells in culture (51) and in a preliminary study on human tumours (159). Many of these results may be accounted for by the dependence of the to tal SOD activity on the activity of Mn SOD, and hence a reduction of Mn SOD results in a reduction of to ta l SOD. In the light of the present studies, however, a reduction of to tal SOD is not a characteristic of human tumours.

C atalase activities

The catalase activities varied a lot in tissues, the highest activity being in the liver and in the erythrocyte which had almost 1.500 U/mg protein. Low activities were found in several other organs, for example the heart with

54 U/mg, and the spleen had a similar level. Even lower activities were found in skeletal muscle, brain grey and white m atter and thyroid gland. Interm ediate activities were found in kidney, adrenal, lung, pancreas and lymphatic node. Interestingly adipose tissue contained also interm ediate catalase activities. Catalase levels were not determined in the tumours.

Catalase activities in the cell lines also varied considerably and were

largest in the HL60 promyleocytic leukemia cell line with an enzyme activity of 920 U/mg protein, and were also high in the histiocytic lymphoma U937, the KG1 myeloblastic cell line and in HeLa. The lowest

- 49 -

catalase activities were found in the endothelial cell, EC, which is not neoplastic, with an enzyme activity of 5.0 U, and in the mesothelioma cell line P31 with 5.5 U. The mean catalase activity of the 27 cell lines which were analyzed for this enzyme was 125 U/mg protein. Due to the skewed distribution of the activities, most of the activities centered around 50 U/mg protein; the median value for the catalase activity in the myeloma cell line RPMI 8226 however being 67 U. Catalase is also known to be an inducible enzyme in several systems (80,119) which may explain the variation in the enzyme activities.

In the light of the present results, the tissues with the highest metabolism again seem to have the highest catalase activities. Among the cell lines most neoplastic cells appeared to have little catalase. Reduced catalase activities have also been reported in other studies on tumours (13,43,119). In conclusion the present study, however, failed to show a general correlation between catalase activities in tissues or tumour cells and assumed radiosensitivity. Furtherm ore no relationship between catalase activities and the neoplastic s ta te 'per se' was found.

Glutathione peroxidase activities

In tissues, glutathione peroxidase activities (GSHPX) were comparatively more evenly distributed than the other enzymes with the highest activity in the liver (about 10,000 U/g wet weight or 150-200 U/mg protein). The difference between the liver and the other tissues was not as great as it was in the case of catalase and SOD; indeed several tissues as spleen ,renal cortex and erythrocytes contained almost the same activity. Several other tissues e.g. adrenal gland, renal medulla, pancreas, lymphatic node, lung, heart, brain and adipose tissue had values about half or more of the liver enzyme activity. The glutathione peroxidase activities in the tumours were not determined.

The liver also contains a large amount of glutathione-S-transferase enzyme which possesses glutathione peroxidase activity (139), and this enzyme activity was not assessed by the present assay. The occurrence of this

- 50-

transferase in the liver, thereby in reality raises the glutathione peroxidase activity of the liver above tha t of other organs.

Glutathione peroxidase activities were also more evenly distributed in the cell lines. The highest GSHPX activities were found in the U937 histiocytic lymphoma cell line and in the KG1 myeloblastic cell line which had enzyme activity of 250 U. The lowest activity was found in the Burkitt lymphoma cell line Namalwa with 1.5 U/mg protein. The mean value of 26 cell lines was 54 U, with an SEM of 11 U (20%). Most values centered around an enzyme activity of 50 U/mg protein.

Glutathione peroxidase removes not only hydrogen peroxide but also other hydroperoxides, and thus has a broader protective ability than catalase.

The enzyme was discovered rather recently and investigations on the enzyme activities in human neoplastic cells have not yet been presented. Reduced activities of glutathione peroxidase have been reported in mouse and ra t transplantable tumours (25,133).

Glutathione peroxidase activities in neoplastic cells were thus within the range of normal cells and tissues, and no relationship was detected between glutathione peroxidase activity and assumed radiosensitivity of tissues and cells.

Hydrogen peroxide accumulation

Rotilio e t al. (25,26,147,), have proposed tha t increased ratios of SOD to catalase or preferably SOD to glutathione peroxidase might be character­istic of some tumour cells. The significance of this would be tha t the decreased ability of the neoplastic cells to take care of hydrogen peroxide, coupled to an increased relative amount of SOD which generates hydrogen peroxide, could increase their sensitivity to hydrogen peroxide. In parti­cular the sensitivity of such cells to increased levels of hydrogen peroxide, produced by irradiation or treatm ent with certain antitumour drugs, would be increased, and possibly influence their sensitivity towards these agents. In the present study a high ratio of SOD to GSHPX, or SOD to catalase was

- 51-

present in some cells but it was not a general feature. The mesothelioma P31 ratios were both rather high, and in other cell lines there was a simultaneous occurrence of a low ratio of SOD to catalase and a high ratio of SOD to GSHPX, e.g. CCRF-CEM with a SOD/catalase ratio of 1.1 and a SOD/GSHPX ratio of about 40, and Namalwa with the ratios 0.6 and 141 respectively. On basis of this, the proposition of Rotiiio is not contradicted by the present study, but it seems that it does not apply to all tumour cells.

Decreased ratios of hydrogen peroxide generating-enzymes to hydrogen peroxide-degrading enzymes was also looked for. However decreased SOD/catalase or SOD/GSHPX ratios was not a general characteristic of the tumour cell lines compared to the few normal cell lines (adult skin, embryonal lung, and endothelial cell line EC). Some tumour cell lines contained rather low ratios, e.g. U937 with a SOD/catalase ratio of 0.5 and a SOD/GSHPX ratio of 1.2, and HL60 with the corresponding ratios of 0.25 and 3.5 respectively.

In no case was there a low ratio of SOD/catalase in combination with a low SOD/GSHPX ratio. This indicates tha t in the addition to SOD the cell must contain a t least one of the catalase degrading enzymes to be able to survive.

Free radical producing cytotoxic drugs

Although the principal aim of the Thesis was to answer the question of the importance of SOD for radiosensitivity, it is also evident that no simple, obvious relationship was observed for the enzyme activités in tumours or tumour cells and their assumed sensitivity to radical-producing antitumour agents.

The findings in heart tissue of low activities of catalase and glutahione peroxidase however, are in accordance with previous reports (45), and in fac t the present measurements are more relevant as the enzymes were determined not only in the heart and the liver but in most tissues. The low activities of these enzymes may, therefore, contribute to the suscepti­

- 52-

bility of the heart to be damaged by Adriamycin or other free radical producing anti-tum our drugs.

Conclusions

The activities of superoxide dismutase, catalase and glutathione peroxidase were determined in human tissues, human tumours (only superoxide dismutase) and human cell-lines. The metabolically most active tissues possessed the highest enzyme activités. In all of the investigated systems the variation of enzyme activités was large for Mn superoxide dismutase and catalase, and less for CuZn superoxide dismutase and glutathione peroxidase. No obvious correlation was found between superoxide dismutase activity and assumed radiation sensitivity.

- 53-

3 CONCLUSIONS

1. Treatm ent of Chinese hamster V 79 cells or erythrocytes with diethyl- dithiocarbam ate reduced substantially their CuZn SOD activity.

2. The DDTC-induced reduction of intracellular CuZn SOD increased the radiation damage (clonogenic survival) of proliferating, metabolically active, V 79 cells through a post-irradiation mechanism related to potentially lethal damage. However, DDTC-induced reduction of in tra­cellular CuZn SOD did not modify the radiation damage (haemolysis) of non-proliferating, metabolically arrested erythrocytes, and no reduction in radiation haemolysis was seen when the extracellular activity of CuZn SOD was increased.

3. The activity levels of CuZn SOD and Mn SOD were determined in normal human tissues, human neoplastic tumours and normal and neoplastic human cell-lines. There was no obvious correlation between

the enzyme activities and the assumed radio-sensitivities of the tissues, tumours and cell-lines.

4. The increased radiation sensitivity seen in DDTC-treated cells was too small to be detected in studies like (II) and (III).

- 54-

4 ACKNOWLEDGEMENTS

I express my sincere gratitude to everyone that have contributed to this study, and I am especially indebted to:

Stefan Marklund, who suggested this study, and have been my principal teacher and collaborator. Without his extensive knowledge of free radicals and superoxide dismutase, his enthusiasm, patience and constructive criticism this study would have been impossible.

Bo Littbrand, professor of Oncology, for his im portant advice on radiobioloical issues, and rewarding discussions during the investigation.

Göran Roos, Erik Lundgren, Jörgen Midander and Mikael Sjöström, for

teaching me the a rt of cell culture, something about flow cytofluorometry and electron microscopy respectively and for their collaboration in the production of this Thesis and publication (III) and (V).

Agneta Öberg, B irgitta Nilsson, Kjerstin Lundmark, Tord Johansson,

Ingegerd Hedlöf, Rut Jonsson and Margareta Bäckström for skilful technical assistance.

Kerstin Hägg, Gunilla Lundmark, Kristina Lindberg for typing, editing and helping in adm inistrative problems.

The staffs a t the Dept, of Surgery, the Dept, of Radiotherapy, and the Depts. of Pathology and Cytology, for help with obtaining tissue and tumour samples, and giving access to the radiation treatm ent machines.

All my friends a t the Department of Oncology.

The Radiobiological Department and Christina Bengzon, report centre a t FOA 4, Umeå.

And Kerstin, Stina and Olle.

This investigation was supported by grants from the Swedish Cancer Society and the Lions Research Foundation, Department of Oncology, Umeå, Sweden.

- 55-

5 REFERENCES

1. Abe, M., Nishiai, T., Yukawa, Y., Takahashi, M., Ono, K., Hiraoka, M., Ri, N. Studies on the radioprotective effects of superoxide dismutase in mice. Int. 3. Radiat. Oncol. Biol. Phys. 1981;7:205-209.

2. Ala-Ketola, L., Varis, R., Kiviniitty, K. E ffect of ascorbic acid on the survival of rats after whole body irradiation. Strahlentherapie,1974;148:643-644.

3. Alexander, P., Charlesby, A. Physico-chemical methods of protection against ionizing radiation. In. ’Radiobiology Symposium1, Liege 1954, p.49. Edited by Bacq, Z.M., Alexander, P., Butterworth, London,

1955.

4. Alexander, P., Bacq, Z.M., Cousens, S.F., Fox, M., Herve, A., Lazar,3. Mode of action of some substances which protect against th eJe th a leffects of X-rays. Radiat. Res. 1955;2:392-415.

5. Altman, K.I., Gerber, G.B., Okada, S. Radiation Biochemistry,Ac.Press, New York, USA, 1970, Vol. I., pp. 78-79.

6. Asada, K., Takahashi, M., Tanaka, K., Nakano, Y. Formation of

active oxygen and its fa te in chloroplasts. In: Biochemical andmedical aspects of active oxygen. Hayashi, O., Asada, K. (editors), University Park, Baltimore, New York, 1977, pp. 45-63.

7. Asano, M., Wakabayashi, T., Kurono, C., Ozawa, T. Mechanism of theformation of megamitochondria induced by copper chelating agents.III. Formation and some biochemical properties of megamitochondria produced by diethyldithiocarbam ate. Acta Path. 3ap. 1975;25:125- 134.

8. Aust, S.D., Roerig, D.L., Pedersen, T.C. Evidence for superoxide generation by NADPH-cytochrome c reductase of ra t liver micro- somes. Biochem. Biophys. Res. Commun. 1972;47:1133-1137.

- 56-

9. Aust, S.D., Svingen, B.A. The role of iron in enzymatic lipid peroxida­tion. In: Free Radicals in Biology, Vol. V., Pryor, W.A. (editor), Ac. Press, New York, USA, 1982, pp. 1-29.

10. Babior, B.M. Oxygen dependent microbial killing by phagocytes. Part I. New Engl. 3. Med. 1978;298:659-668.

11. Bachur, N.R., Gordon, S.L., Gee, M.V. A general mechanism for microsomal activation of quinone anticancer agents to free radicals. Cancer Res. 1978;38:1745-1750.

12. Bacq, Z.M., Herve, A. Action radioprotectrice des bloqueurs de cuivre (agents de chelation). Arch.Internat.Physiol. 1953; Vol. LXI., Fase.3:433-434.

13. Barkowiak, A., Barkowiak, 3. Superoxide dismutase and catalase activities in normal and cancerous tissue. Comp. Biochem. Physiol. 1981;70B:819-820.

14. Bartoli, G.M., Galeotti, T., Azzi, A. Production of superoxide anions and hydrogen peroxide in Ehrlich ascites tumour cell nuclei. Biochim. Biophys. Acta 1977;497:622-626.

15. Bartoli, G.M., Bartoli, S., Galeotti, T., Bertoli, E. Superoxide

dismutase content and microsomal lipid composition of tumours with different growth rates. Biochim. Biophys. Acta 1980;620:205-211.

16. Bartosz, G., Leyko, W. Radioprotection of bovine erythrocytes to haemolysis. Int. 3. Radiat. Biol. 1981;39:39-46.

17. Beaumariage, M.L. Essais de prevention e t traitm em ent du Mal des Rayons chez le Poussin. Int. 3. Appi. Radiat. Isotopes 1958;4:69-83.

18. Behar, D., Czapski, G., Dorfman, L.M., Rabani, 3., Schwarz, H.A. The

acid dissociation constant and decay kinetics of the perhydroxyl radical. 3. Phys. Chem. 1970;74:3209-3213.

- 57 -

19. Bert, P. La Pression Barométrique, Paris, 1878. (Translated by M.A. Hitchcock and F.A. Hitchcock.'Barometric Pressure: Researches in Experimental Physiology1, College Book Co., Columbus, Ohio, USA,

1943.)

20. Bize, I.B., Oberley, L.W., Morris, H.P. Superoxide dismutase and Superoxide radical in Morris hepatomas. Cancer Res. 1980;40:3686- 3693.

21. Bode, H. Systematische Untersuchungen über die Anwendbarkeit der Diäthyldithiocarbaminate in der Analyse. Z.Anal.Chem., 19 54; Bd.142:414-426. idem, Bd.l43:182-195. idem, Bd. 144:165-186.

22. Borch, R.F., Pleasants, M.E. Inhibiton of cis-platinum nephrotoxicity

by diethyldithiocarbam ate rescue in a ra t model. Proc. Natl. Acad. Sci. USA 1979;76: 6611-6614.

23. Borch, R.F., Katz, J.C ., Lieder, P .H., Pleasants, M.E. E ffect of

diethyldithiocarbam ate rescue on tumor response to cis-platinum in a ra t model. Proc. Natl. Acad. Sci. USA, 1980;77:5441-5444.

24. Bors, W., Saran, M., Czapski, G. The nature of interm ediates during biological oxygen activation. In: Biological and Clinical Aspects of Superoxide and Superoxide Dismutase. Bannister, W.H. and Bannister, J.V. (editors), Elsevier/North Holland, New York, USA, 1980, pp. 1-31.

25. Bozzi, A., Mavelli, I., Finazzi Agro, A., Strom, R., Wolf, A.M.,

Mondovi, B., Rotilio, G. Enzyme defense against reactive oxygen derivatives. II. Erythrocytes and tumor cells. Mol. Cell Biochem. Physiol. 1976; 10/1:11—16.

26. Bozzi, A., Mavelli, I., Mondovi, B., Strom, R., Rotilio, G. D ifferential sensitivity of tumor cells to externally generated hydrogen. Role of glutathione and related enzymes. Cancer Biochem. Biophys. 1979;3:135-141.

- 58-

27. Bradford, M.M. A rapid and sensitive method for the quantitation of protein utilizing the principle of protein dye binding. Anal.Biochem. 1976;10:11-16.

28. Brannan, T.S., Maker, H.S., Bernstein, M. Postmortem stability of enzymes detoxifying peroxide in brain. J.Neurochem. 1982;39:589- 591.

29. Cameron, G.R., Rydon, H.N., Porton Report 1944.

30. Capei, I.D., Thornley, A.C. Superoxide dismutase activity, caerulo­plasmin activity and lipoperoxide levels in tumour and host tissues of mice bearing the Lewis lung carcinoma. Eur.J.Cancer Clin.Oncol.

1982;18:507-513.

31. C asarett, A.P. Radiation Biology, Prentice Hall, New Jersey, USA, 1968. pp. 57-89.

32. Chance, B., Sies, H.,Boveris, A. Hydroperoxide metabolism in

mammalian organs. Physiol. Rev. 1979;59:527-605.

33. Chapman, J.D ., Reuvers, A.P., Borsa, J., Greenstock, C.L. Chemical

radioprotection and radiosensitization of mammalian cells grown in vitro. Radiat. Res. 1973;56:291-306.

34. Cocco, D., Calabrese, L., Rigo, A., Argese, E., Rotilio, G. Re­examination of the reaction of diethyldithiocarbam ate with the copper of superoxide dismutase. J. Biol. Chem. 1981;256:8983-8986.

35. Cohen, G., Heikkila, R.E. The generation of hydrogen peroxide,

superoxide radical and hydroxyl radical by 6-hydroxydopamine, dialuric acid and related cytotoxic agents. J. Biol. Chem. 1974; 249:2447-2452.

- 59-

36. Cohen, G., Sinet, P. Fenton’s reagent - once more revisited. In: Chemical and Biochemical Aspects of Superoxide and Superoxide Dismutase. Bannister, J.V. and Hill, H.A.O. (editors), Elsevier/North Holland, New York, USA, 1980, pp. 27-37.

37. Del Rio, L., Ortega, M.G., Lopez, A.L., Gorge, 3.L. A more sensitive modification of the catalase assay with the Clark oxygen electrode. Anal. Biochem. 1977;80:409-415.

38. DeLuca, L.M., Shapiro, S.S. (editors) Modulation of cellular in ter­actions by vitamin A and derivatives (retinoids). Ann. N.Y. Acad. Sci. USA. 1981 ;359.

39. Demopoulos, H.B. The basis of free radical pathology. Fed. Proc. 1973;32:1859-1861.

40. Deneke, S.M., Fanburg, B.L. Involvement of glutathione enzymes in

O2 tolerance development by diethyldithiocarbam ate. Biochem. Pharmacol. 1980;29:1367-1373.

41. Dimond, A.E., Horsfall, J.G ., Heuberger, J.W., Stoddard, E.M. Role of the dosage-response curve in the evaluation of fungicides. Conn. Exp. S tate Bull. 1941;451:635-667.

42. Dionisi, O., G aleotti, T., Terranova, T., Azzi, A. Superoxide radicals and hydrogen peroxide formation in mitochondria from normal and neoplastic tissue. Biochim. Biophys. Acta 1975;403:292-300.

43. Dionisi, O., G aleotti, T., Terranova, T., Azzi, A. Superoxide radicals and hydrogen peroxide formation in mitochondria from normal and neoplastic tissue. Biochim. Biophys. Acta 1975;403:292-300.

44. Docampo, R., Cruz, F.S., Boveris, A., Moniz, P.A., Esquivel, D.M. ß- lapachone enhancement of lipid peroxidation and superoxide anion and hydrogen peroxide formation by Sarcoma 180 ascites tumor cells. Biochem. Pharmacol. 1979;28:723-728.

- 60-

45. Doroshow, 3.H., Locker, G.Y., Myers, C.E., Enzymatic defences of the mouse heart against reactive oxygen m etabolites. 3.Clin.Invest. 1980;65:128-135.

46. Draganic, I.G., Draganic, Z.D. Radiation Chemistry of Water. Ac.Press, New York, USA, 1971, p.140.

47. Dryer, S.E., Dryer, R.L., Autor, A.P. Enhancement of mitochondrial, cyanide resistant superoxide dismutase in the livers of ra ts treated with 2,4-dinitrophenol. J.Biol.Chem. 1980;255:1054-1057.

48. Eikind, M.M., Sutton, H. Radiation response of mammalian cells grown in culture. I. Repair of X-ray damage in surviving Chinese hamster cells. Radiat.Res. 1960;13:556-593.

49. Emerit, I. Chromosome breakage factors: origin and possible signifi­

cance. In: Progress in Mutation Research, Voi 4., Natarajan, A.T. e t al. (editors), Elsevier Biomedical Press, 1982, p. 61-74.

50. Fee, 3.A. Chemical and physical properties of superoxide. In: Super­oxide and superoxide dismutases. Michelson, A.M., McCord, 3 .M., Fridovich, I. (editors), Ac. Press, New York, USA, 1977, pp. 19-60.

51. Fernandez-Pol, 3.A., Hamilton, P.D., Klos, D.3. Correlation between the loss of the transformed phenotype and an increase in superoxide dismutase activity in a revertant subclone of sarcoma virus-infected mammalian cells. Cancer Res. 1982;42:609-617.

52. Flohe, L., Günzler, W.A., Schock, H.H., Glutathione peroxidase: a seleno-enzyme. FEBS L ett. 1973;32:132-134.

53. Flohe, L. Glutathione peroxidase brought into focus. In: 'Free radicals

in Biology1, Vol. V., Pryor, W.A. (editor), Ac. Press, New York, USA, 1982, pp. 223-275.

- 61-

54. Frank, L., Wood, D.L., Roberts, R .J. E ffect of diethyldithiocarba- m ate on oxygen toxicity and lung enzyme activity in immature and adult rats. Biochem. Pharmacol. 1978;27:251-254.

55. Fridovich, I. The biology of oxygen radicals. Science, 1978;201:875- 880.

56. Fried, R. Enzymatic oxidation of diethyldithiocarbam ate by xanthine oxidase and its colorim etric assay. Ann.N.Y.Acad.Sci. 1976;273:212- 218.

57. Gale, G.R., Smith, A.B., Walker, E.M. Diethyldithiocarbamate in treatm en t of acute cadmium poisoning. Ann. Clin. Lab. Sci.

1981;11:476-483.

58. G aleotti, T., Borello, S., Seccia, A., Faralo, E., Bartoli, G.M., Serri, F. Superoxide dismutase content in human epidermis and squamous cell epithelioma. Arch. Dermatol. Res. 1980;267:83-86.

59. G aleotti, T., Bartoli, G.M.,Bartoli, S., Bertoli, E. Superoxide radicals and lipid peroxidation in tumour microsomal proteins. In: Biological and Clinical Aspects of Superoxide and Superoxide Dismutase. Bannister, W.H. and Bannister, J.V. (editors), Elsevier/North Holland, New York, 1980, pp. 106-117.

60. Gallagher, C.H., Reeve, V.E. Interrelationships of copper, cyto­chrome oxidase, phospholipid synthesis and adenine nucleotide binding. Aust.J.Exp. Biol.Med.Sci. 1976;54:593-600.

61. Geiss, D., Schulze, H.U. Isolation and chemical composition of the

NADH:semidehydroascorbate oxidoreductase-rich membranes from ra t liver. FEBS L ett. 1975;60:374-379.

62. Gilbert, D.L. (editor) Oxygen and living processes. An interdisciplin­ary approach. Springer-Verlag, New York, USA, 1981.

- 62-

63. Gilman, S.C., Fischer, G.3. The e ffec t of diethyldithiocarbam ate on passive avoidance learning by chicks. Psychopharmacol. 1976;48:287- 289.

64. Goldstein, B.D., Rozen, M.G., Quintavalla, J.C ., Amoruso, M.A.Decrease in mouse lung and liver glutathione peroxidase activity and potentiation of the lethal e ffec ts of ozone and paraquat by the superoxide dismutase inhibitor diethyldithiocarbam ate. Biochem.Pharmacol. 1979;28:27-30.

65. Goh, K.O., Sumner, H. Breaks in normal human chromosomes: arethey induced by a transferable substance in the plasma of personsexposed to to tal body irradiation? Radiat.Res. 1968;35:171-174.

66. Gottfried, E.L., Robertson, N.A. Glycerol lysis time as a screening test for erythrocyte disorders. J.Lab.Clin.Med. 1974;83:323-333.

67. Graf ström, R., Greene, F.E. Differential e ffects of disulfiram and diethyldithiocarbam ate on small intestinal and liver microsomal benzo/a/pyrene metabolism. Biochem.Pharmacol. 1980;29:1517-1523.

68. Grankvist, K., Marklund, S., Täljedahl, I-B. Superoxide dismutase is prophylactic against alloxane diabetes. Nature, 1980;294; 158-160.

69. Gray, B.H., Stull, R. Polyethylene Glycol Enzyme Complexes as radioprotectors. Radiat.Res 1982,91,400. (Abstract).

70. Gregory, E.M., Fridovich, I. Induction of superoxide dismutase by molecular oxygen. 3.Bact. 1973;114:543-548.

71. Gregory, E.M., Goscin, S.A., Fridovich, I. Superoxide dismutase and

oxygen toxicity in a eukaryote. J.Bact. 1974; 117:456-460.

72. Griffiths, D.E., Wharton, D.C. Studies on the electron transport system. XXXV. Purification and properties of cytochrome oxidase. 3.Biol.Chem. 1961;236:1850-1856.

- 63-

73. Guarnieri, C., Flamigni, F., Ventura, C., Rossoni-Caldarera, C. Inhibition of ra t heart superoxide dismutase activity by diethyldithio- carbam ate and its e ffec t on mitochondrial function. Biochem. Pharmacol. 1981;30:2174-2176.

74. Günzler, W.A., Kremers, H., Flohe, L. An improved coupled te s t procedure for glutathione peroxidase in blood. Z.Klin. Chem. Klin. Biochem. 1974;12:444-448.

75. Haber, F., Weiss, 3. The cataly tic decomposition of hydrogen peroxide by iron salts. Proc. Roy. Soc. Lond. 1934;A147:332-351.

76. Hall, E.3. Radiobiology for the radiologist (second edition).

Harper&Row Publish. Inc., New York, USA, 1978, p. 122.

77. Halliwell, B., Gutteridge, 3.M.C. Formation of a thiobarbituric acid reactive substance from deoxyribose in the presence of iron salts. FEBS Lett. 1981;128:347-352.

78. Halliwell, B. Superoxide and superoxide-dependent formation of hydroxyl radicals are im portant in oxygen toxicity. Trends Biochem. Sci., August 1982:270-272.

79. Harman, D. The free radical theory of aging. In: Free Radicals in Biology, Vol.V., Pryor, W.A. (editor), Ac. Press, New York, USA, 1982, p. 255-275.

80. Hassan, H.M., Fridovich, I. Intracellular production of superoxide radicals and hydrogen peroxide by redox active compounds. Arch. Biochem. Biophys. 1979;196:385-395.

81. Haugaard, N. Mechanisms of oxygen toxicity. Physiol. Rev. 1968;48:311-373.

- 64-

82. Hedenstedt, A., Rannug, U., Ramel, C., W achtmeister, C.A. Muta­genicity and metabolism studies on 12 thiuram and dithiocarbam ate compounds used as accelerators in the Swedish rubber industry. Mut.Res. 1979;68:313-323.

83. Heikkila, R.E., Cabbat, F.S., Cohen, G. In vivo inhibition of super­oxide dismutase in mice by diethyldithiocarbam ate. 3. Biol. Chem. 1976;231:2182-2183.

84. Heikkila, R.E., Cabbat, F.S., Cohen, G. Inactivation of superoxide dismutase by several thiocarbamic acid derivatives. Experientia 1978;34/12: 1333-1334.

83. Hodgson, E.K., Fridovich, I. The interaction of bovine erythrocytesuperoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry 1973;14:3294-3299.

86. Hollowell, 3.G., L ittlefield, L.G. Chromosome damage induced by plasma of X-rayed patients. Proc. Soc. Exp. Biol. Med. 1968; 129:240- 243.

87. Huber, W. Orgotein - (Bovine CuZn Superoxide Dismutase), an anti­

inflammatory protein. Discovery, toxicology and pharmacology. Eur. 3. Rheum. Inflam. 1981;4:173-182.

88. Ingold, 3.A., Reed, G.B., Kaplan,H.S., Bagshaw, M.A. Radiation

hepatitis. Am. 3. Roentgenol. 1963;93:200-208.

89. 3ensen, 3.C., Faiman, M.D. Determination of disulfiram and m etabolites from biological fluids by high performance liquid chromatography. Chromatography 1980;181:407-416.

90. 3irtle, R.L., McLain, 3.R., Strom, C.S., Michalopoulos, G. Repair of radiation damage in noncycling parenchymal hepatocytes. Br. 3. Radiol. 1982;33,847-831.

- 65-

91. Kosower, N.S., Kosower, E.M. The glutathione-glutathione disulfide system. In: Free Radicals in Biology, Vol. II., Pryor, W.A. (editor), Ac. Press, New York, USA, 1976, pp. 55-84.

92. Lang, M., Marselos, M., Törönen, R. Modification of drug metabolismby disulfiram and diethyldithiocarbam ate. I. Mixed-functionoxygenase. Chem.-Biol. Interact. 1976;15:267-276.

93. Lengfelder, E. On the action of diethyldithiocarbam ate as inhibitor of copper-zinc superoxide dismutase. Z.Naturforsch. 1979;34c:1292- 1294.

94. Levinson, W., Mikelens, P., Opperman, H., 3ackson, 3. Effect ofAntabuse (Disulfiram) on Rous sarcoma virus and on eukaryotic cells.Biochim. Biophys.

95. Lin, P-5., Kwock, L., Butterfield, C.E. Diethyldithiocarbamate

enhancement of radiation and hyperthermic effects on Chinese ham ster cells in vitro. Radiat. Res. 1979;77:501-511.

96. Lin, P-S., Kwock, L., Lui, P., H efter, K. Enhancement of oxygen

toxicity by diethyldithiocarbam ate on Chinese ham ster and mouse cells. Int. 3. Radiat. Oncol. Biol. Phys. 1979;5:1699-1703.

97. Lin, P.S., Kwock, L., Goodchild, N.T. Copper chelator enhancement of Bleomycin cytotoxicity. Cancer 1980;46:2360-2364.

98. Littbrand, B., Revesz, L., The e ffec t of oxygen on cellular survival and recovery a fte r radiation. Brit. 3. Radiol. 1969;42:914-924.

99. Loven, D.P., Oberley, L.W., Rousseau, F.M., Stevens, R.H. Superoxide dismutase activity in 1,2-dimethylhydrazine-induced ra t colon adeno­carcinoma. 3. Natl. Cancer Inst. 1980;65:377-381.

- 66-

100. Lutz, L.M., Glende 3r, E.A., Recknagel, R.O. Protection by diethyl- dithiocarbam ate against carbon tetrachloride lethality in rats and against carbon tetrachloride induced lipid peroxidation in vitro. Biochem. Pharmacol. 1973; 22:1729-1734.

101. Mann, T., Keilin, D. Haemocuprein and hepatocuprein, copper-protein compounds of blood and liver in animals. Proc. Roy. Soc.(London), 1938;B 126:303-315.

102. Marklund, S.L. Spectrophotometric study of spontaneous dispro­portionation of superoxide anion radical and sensitive direct assay for superoxide dismutase. 3. Biol. Chem. 1976;251:2182-2185.

103. Marklund, S.L. A simple specific method for the determ ination of the

hemoglobin content of tissue homogenates. Clin. Chim. Acta 1979; 92:229-234.

104. Marklund, S.L., Westman, G. CuZn superoxide dismutase and the

radiation sensitivity of Chinese hamster cells. In: Biological and Clinical Aspects of Superoxide and Superoxide Dismutase. Bannister,3.V. and Bannister, W,H. (editors), Elsevier/North Holland, New York Amsterdam, 1980, pp. 318-326.

105. Marklund, S.L. Human copper-containing superoxide dismutase of

high molecular weight. Proc. Natl. Acad. Sci. USA 1982; 79: 7634-7638.

106. Marselos, M., Lang, M., Törrönen, R. Modifications of drug m eta­bolism by disulfiram and diethyldithiocarbam ate. II. D-glucuronic acid pathway. Chem.-Biol. Interact. 1976;15:277-287.

107. Mason, R.P., Free radical interm ediates in the metabolism of toxic chemicals. In: Free Radicals in Biology, Vol. V., Pryor, W.A. (editor), Ac. Press, New York, USA, 1982, pp. 161-221.

- 67-

108. McBrien, D.C.H., Slater, T.F. (editors) Free Radicals, Lipid Peroxida­tion and Cancer. Ac. Press, New York, USA, 1982.

109. McCord, 3.M., Fridovich, I. Superoxide dismutase. An enzymatic function for erythrocuprein (hemocuprein). 3. Biol. Chem. 1969; 244: 6049-6055.

110. McCord, 3.M., Keele, B.B., Fridovich, I. An enzyme based theory of obligate anaerobiosis: The physiologic function of superoxide dismutase. Proc. Natl. Acad. Sci. USA 1971;68:1024-1027.

111. McCord, 3.M. Free radicals and inflammation: protection of synovial fluid by superoxide dismutase. Science, Wash. D.C. 1974;185:529-531

112. Michael, B.D., Adams, G.E., Hewitt,H.B., 3ones, W.B.G., Watts, M.E. A posteffect of oxygen in irradiated bacteria: a submillisecond fast mixing study. Radiat. Res. 1973;54:239-251.

113. Michelson,A.M., Buckinham, M.E. Effects of superoxide radicals on

myoblast growth and differentiation. Biochem. Biophys. Res. Commun. 1974;58:1079-1086.

114. Midander, 3., Deschavanne, P.3., Malaise, E.P., Revesz, L. Survival curves of irradiated glutathione deficient human fibroblasts: Indication of a reduced enhancement of radiosensitivity by oxygen and misonidazole. Int. 3. Radiat. Oncol. Biol. Phys. 1982;8:443-446.

115. Miller, R.C., Osmak, R., Zimmerman, M., Hall, E.3. Sensitizers,

protectors and oncogenic transform ation in vitro. Int. 3. Radiat. Oncol. Biol. Phys. 1982;8:771-775.

116. Mills, G.C. Hemoglobin catabolism, I. Glutathione peroxidase, an

erythrocyte enzyme which protects hemoglobin from oxidative breakdown. 3. Biol. Chem. 1957;228:6084-6090.

- 68-

117. Misra, H.P., Fridovich, I. Superoxide dismutase and the oxygen enhancement of radiation lethality. Arch. Biochem. Biophys. 1976; 176:577-581.

118. Misra, H.P. Reaction of copper-zinc superoxide dismutase with diethyl-dithiocarbam ate. 3. Biol. Chem. 1979;254:11623-11628.

119. Mochizuki, Y., Hruban, Z., Morris, H.P., Slesers, A., Vigil, E.L. Microbodies of Morris hepatomas. Cancer Res. 1971;31:753-773.

120. Neveu, P.3. The effec ts of thiol moiety of levamisole on both cellular and humoral immunity during early response to a hapten-carrier complex. Clin. Exp. Immunol. 1978;32:419-422.

121. Newton, G.L., Dorian, R., Fahey, R.C. Analysis of biological thiols: derivatization with monobromobimane and separation by reverse- phase high performance liquid chromatography. Anal. Biochem. 1981; 114:383-387.

122. Nordenson, I. E ffects of superoxide dismutase and catalase on radiation induced chromosome abberations: dose and cell cycle dependence. Hereditas 1978; 89:163-167.

123. Nordenson, I., Beckman, G., Beckman, L., Nordstrom, S. The effec t

of superoxide dismutase and catalase on radiation-induced chromo­some breaks. Hereditas 1976;82:125-126.

124. Oberley, L.W., Lindgren, A.L., Baker, S.A., Stefens, R.H. Superoxide ion as the cause of the oxygen effec t. Radiat. Res. 1976;68:320-328.

125. Oberley, L.W., Buettner, G.R. Role of superoxide dismutase in cancer: A review. Cancer Res. 1979;39:1141-1149.

126. Oberley, L.W., Bize, I.B., Sahu, S.K., Leuthauser, S.W.H., Gruber, H.E. Superoxide dismutase activity of normal murine liver,

- 69-

regenerating liver, and H6 hepatoma. 3. Natl. Cancer Inst. 1978; 61:

375-379.

127. Oberley, L.W., Sahu, S.K., Cheng, F.H.F. Superoxide dismutase activity in normal mouse brain, prenatal brain, and neuroblastoma cells. AACR 1978, A bstract No 586.

128. Oberley, L.W., Oberley, T.D., Buettner, G.R. Cell division in normal and transformed cells: the possible role of superoxide and hydrogen peroxide. Medical Hypotheses, 1981;7:21-42.

129. O'Connor, M.K., Malone, 3.F., Moriarty, M., Mulgrew, S. A radio- protective e ffec t of vitamin C observed in Chinese hamster ovary

cells. Brit. 3. Radiol. 1977;50:587-591.

130. Oshino, N., Chance, B., Sies, H., Biicher, T. The role of hydrogen

peroxide generation in perfused ra t liver and the reaction of catalase compound I and hydrogen donors. Arch. Biochem. Biophys. 1973; 154: 117-131.

131. Overgaard, 3., Nielsen, O.S., Overgaard, M., Steenholdt, S., 3akobsen, A., Sell, A. Lack of radiation protective effec t of orgotein in normal and malignant mammalian cells. Acta Radiol. Oncol. 1979; 18:305- 312.

132. Parks, D.A., Bulkley, G.B., Granger, D.N., Hamilton, S.R., McCord,3.M. Ischemic injury in the ca t small intestine: role of superoxide radicals. Gastroenterology 1982;82:9-15.

133. Peskin, A.V., Koen, Y.M., Zbarsky, I.B., Konstantinov, A.A.

Superoxide dismutase and glutathione peroxidase activities in tumours. FEBS Lett. 1978;78:41-45.

134. Petkau, A., Chelack, W.S. Radioprotection of Acholeplasma laidlawii B by superoxide dismutase. Int. 3. Radiat. Biol. 1974;26;421-426.

- 70-

135. Petkau, A., Chelack, W.S., Pleskach, S.D., Meeker, B.E., Brady, C.M. Radioprotection of mice by superoxide dismutase. Biochem. Biophys. Res. Commun. 1973;65:886-893.

136. Petkau, A., Chelack, W.S. Radioprotective effec t of superoxide dismutase on model phospholipid membranes. Biochim. Biophys. Acta 1976;433:445-456.

137. Petkau, A., Chelack, W.S., Pleskach, S.D. Protection of postirradi­ated mice by superoxide dismutase. Int. 3. Radiat. Biol. 1976;29:297- 299.

138. Phillips, R.A., Tolmach, L.3. Repair of potentially lethal damage in X-irradiated HeLa cells. Radiat. Res. 1966;29:413-422.

139. Pierce, S., Tappel, A.L., Glutathione peroxidase activ ities from ra t liver. Biochim. Biophys. Acta 1978;523:27-36.

140. Powell, A.K., E ffect of Dithiocarbamates on sarcoma cells and fibrocytes cultured in vitro. Br. J. Cancer 1954;8:529-534.

141. Reynolds, E.S., Moslen, M.T. Free radical damage in liver. In: Free Radicals in Biology, Vol. IV., Pryor, W.A. (editor), Ac. Press, New York, USA, 1980, pp. 49-94.

142. Rigas, D.A., Eginitis-Rigas, C., Head, C. Biphasic toxicity of diethyl- dithiocarbam ate, a metal chelator, to T lymphocytes and poly­morphonuclear granulocytes: reversal by zinc and copper. Biochem. Biophys. Res. Commun. 1979;88: 373-379.

143. Rigas, D.A., Eginitis-Riga, C., Bigley, R.H., Stankova, L, Head, C. Biphasic radiosensitization of human lymphocytes by diethyldithio- carbam ate; possible involvement of superoxide dismutase. Int. 3. Radiat. Biol. 1980;38: 257-266.

- 71-

144. Roder, 3.C., Helfand, S.L., Werkmeister, 3., McGarry, R., Beaumont, T.3., Duwe, A. Oxygen interm ediates are triggered early in the cytolytic pathway of human NK cells. Nature 1982;298:569-572.

145. Roos, G., Adams, A., Biovanella, B., Lundgren, E., Nilsson, K., Nordenson, I., Sundström, C. Establishment and characterization of a human EBV-negative B cell line (MN 60). Leukemia Res. 1982;6:689- 693.

146. Roots, R., Okada, S. Protection of DNA molecules of culturedmammalian cells from radiation-induced single-strand scission by various alcohols and SH compounds. Int. 3. Radiat. Biol. 1972;21:329- 342.

147. Rotilio, G., Bozzi, A., Mavelli, I., Mondovi, B., Strom, R. Tumor cells

as models for oxygen-dependent cytotoxicity. In: Biological and Clinical Aspects of Superoxide and Superoxide Dismutase, Bannister,

W.H. and Bannister, 3.V. (editors), Elsevier/North Holland, New York, USA, 1980, pp. 118-126.

148. Sahu, S.K., Oberley, L.W., Stevens, R.H., Riley, E.F. Superoxide

dismutase activity of Ehrlich ascites tumor cells. 3. Natl. CancerInst. 1977;58:1125-1128.

149. Samuni, A., Chevion, M., Halpern, Y.S., Czapski, G. Radiation- induced damage in T4 bacteriophage: the e ffec t of superoxide radicals and molecular oxygen. Radiat. Res. 1978;75:489-496.

150. Samuni, A., Chevion, M., Halpern, Y.S., Ilan, Y.A., Czapski, G.

Radiation-induced damage in Escherichia coli B: the effec t of superoxide radicals and molecular oxygen. Radiat. Res. 1978;76:624- 632.

151. Scott, D. The effec t of irradiated plasma on normal chromosomes and its relevance to the long lived lymphocyte hypothesis. Cell Tissue Kinet. 1968;2: 295-298.

- 72-

152. Slater, T.F. Free radical mechanisms in tissue injury. Pion Ltd, London

153. Steinman, H.M., Hill, R.L. Sequence homologies among bacterial and mito-chondrial superoxide dismutases. Proc. Natl. Acad. Sci. USA. 1973; 70:3725-3729.

154. Stjernberg, N., Lundgren, E., Roos, G., Abu Sinna, G. Establishment and characterization of human mesothelioma cell lines. Eur. 3. Respir. Dis. 1982; Suppl. No. 124; Vol63, p.45 (Abstract).

155. Stone, D., Lin, P-S., Kwock, L. Radiosensitization of human erythrocytes by diethyldithiocarbam ate. Int. 3. Radiat. Biol. 1978; 33:393-396.

156. Strömme, 3.H. E ffects of diethyldithiocarbam ate and disulfiram on glucose metabolism and glutathione content of human erythrocytes. Biochem. Pharmacol. 1963;12:705-715.

157. Sunderman, F.W., Clinical response to therapeutic agents in poisoning from mercury vapor. Ann. Clin. Lab. Sci. 1978;8:259-269.

158. Sunderman, F.W. Efficacy of sodium diethyldithiocarbam ate (Dithiocarb) in acute nickel carbonyl poisoning. Ann. Clin. Lab. Sci. 1979;9:1-10.

159. Sykes, 3.A., McCormack, F.X., O'Brien, T.3. A preliminary study of the superoxide dismutase content of some human tumors. Cancer Res. 1978;38: 2759-2762.

160. Szmigielski, A. E ffect of disulfiram and sodium diethyldithiocarba­mate on tyrosine and dopamine-3-hydroxylase activities in ra t brain and heart. Pol. 3. Pharmacol. Pharm. 1975;27:153-159.

- 73-

161. Tappel, A.L. Measurements of and protection from in vivo lipid peroxidation. In: Free Radicals in Biology, Vol. IV., Pryor, W.A. (editor), Ac. Press, New York, USA, 1980, pp. 2-48.

162. Taube, H. Mechanisms of oxidation with oxygen. J. Genet. Physiol. 1963;49:29-52.

163. Thorn, G.D., Ludwig, R.A. The Dithiocarbamates and related compounds. Elsevier, Amsterdam, 1962.

164. den Tonkelaar, E.M., van Logten, M.J. Protective action of dithiocar­bamates on experimental liver damage produced by carbon te tra ­chloride. Toxicol. Appi. Pharmacol. 1974;30:96-106.

163. Totter, 3.R. Spontaneous cancer and its possible relationship to oxygen metabolism. Proc. Natl. Acad. Sci. USA 1980;77:1763-1767.

166. Trush, M.A., Mimnaugh, E.G., Gram, T.E. Activation of pharmaco­

logical agents to radical interm ediates. Biochem. Pharmacol. 1982; 31:3335-3346.

167. Ts’o, P.O.P., Caspary, W.J., Lorentzen, R .J. The involvement of free

radicals in chemical carcinogenesis. In: Free Radicals in Biology, Vol.III., Pryor, W.A. (editor), Ac. Press, New York, USA, 1977, pp. 251- 303.

168. Tyler, D.D. Polarographie assay and intracellular distribution of superoxide dismutase in ra t liver. Biochem. J. 1974; 147:493.504.

169. Van Balgooy, J.N.A., Roberts, E. Superoxide dismutase in normal and malignant tissue in different species. Comp. Biochem. Physiol. 1979: 62B:263-268.

170. Van Bekkum, D.W. The protective action of dithiocarbam ates against

the lethal effects of X-irradiation in mice. Acta Physiol. Pharmacol. Neer. 1956; 4:508-523.

- 74-

171. Van Bekkum, D.W., de Groot, 3. Observations on chemical protection in vivo and in vitro. In: Progress in Radiobiology, Mitchell, 3.G., Holmes, B.E., Smith, C.L. (editors), Oliver & Boyd, Edinburgh, 1956, pp. 243-248.

172. Van Hemmen, 3.3., Meuling, W.3.A. Inactivation of biologically active DNA by y-ray induced superoxide radicals and their dismutation products singlet oxygen and hydrogen peroxide. Biochim. Biophys. Acta 1975;402:133-141.

173. Vindelöv, L.L., Flow microfluorometric analysis of nuclear DNA in cells from solid tumors and cell suspensions. Virchows Arch. B. Cell. Path. 1977;24:227.

174. White, I.G. Studies on the spermicidal activity of chelating agents. Aust. 3. Biol. Sci. 1955;8:387-395.

175. Witting, L.A. The oxidation of a-tocopherol during the autoxidation of ethyl oleate, linoleate, linolenate, and arachidonate. Arch. Bio- chem. Biophys. 1969;129:142-151.

176. Witting, L.A. Vitamin E and lipid antioxidants. In: Free Radicals in Biology, Vol IV., Pryor, W.A. (editor), Ac. Press, New York, USA, 1980, pp. 295-320.

177. Yamanaka, N., Deamer, D. Superoxide dismutase activity in WI-38 cell cultures: e ffec t of age, trypsinization and SV-40 transformation. Physiol. Chem. Phys. 1975;6:95-106.

178. Yamanaka, N., Nishida, K., Ota, K. Increase of superoxide dismutase activity in various human leukemia cells. Physiol. Chem. Phys. 1979; 11:253-256.