THE BIOLOGICAL FATE OF SOME CHLOROFLUOROALKANES

by Peter John Cox

Being a Thesis presented for the award of the Degree of Doctor of Philosophy at the

University of Surrey.

September 1972 Biochemistry Department, University of Surrey, Guildford, Surrey, U.K.

ProQuest Number: 10798359

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a com p le te manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

uestProQuest 10798359

Published by ProQuest LLC(2018). Copyright of the Dissertation is held by the Author.

All rights reserved.This work is protected against unauthorized copying under Title 17, United States C ode

Microform Edition © ProQuest LLC.

ProQuest LLC.789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 48106- 1346

Abstract.

The literature on the toxicology and biochemical effects of chlorofluoroalkanes and closely related compounds of toxicological interest has been reviewed.

The chlorofluoroalkanes used in aerosol propellant mixtures were the subject of this study.

A simple, accurate quantitative g.l.c. assay was developed for the measurement of trichlorofluoromethane and 1 ,2-dichloro-1 ,1 ,2 ,2-tetrafluoroethane in extracts of tissues and the expired air. The separation and quantitation of mixtures of chlorofluoroalkanes was investigated.

The distribution and excretion of trichlorofluoro- methane and 1 ,2 -dichloro-l,1 ,2 ,2-tetrafluoroethane was determined after administration of these compounds by the oral route.

Attempts were made to detect metabolism of trichloro- fluoromethane in in vitro incubations. Although it was feasible to detect 0 .2% of a possible metabolite (dichloro- fluoromethane), no formation of this metabolite was detected in incubations of trichlorofluoromethane with microsomal preparations from several rodent species.Lipid peroxidation was investigated as a measure of free radical formation. Trichlorofluoromethane wasfound to have little effect on this parameter.

Simple toxicological experiments showed the chlorofluoroalkanes to be relatively non-toxic even at

very high dose levels. Total tissue lipid was not affected by trichlorofluoromethane.

The chlorofluoroalkanes gave spectral interactions with cytochrome P-450, both in the oxidised and reduced states. The chlorofluoroalkanes did not affect type I substrate metabolism. The effects on type II substrate metabolism were inhibition (zoxazolamine hydroxylation) and activation (acetanilide and aniline 4-hydroxylations).

The results are discussed in relation to the low toxicity of the chlorofluoroalkanes, and in comparison to the hepatotoxin, carbon tetrachloride.

To my parents and my wife, Janet.

Acknowledgements.

My deepest gratitude is due to Dr. L.J. King for his constant interest and instruction over the past three years; also to Professor D.V. Parke for his enthusiasm and guidance, and for making the facilities of his department available to me.

I would like to thank all the other members of the academic and technical staffs for their help and friendliness; also my fellow postgraduates, particularly Mr. B.G. Lake, for advice on some of the methodology used in this work.

I owe a great deal to my parents, without whose help and encouragement I would never have undertaken this period of training.

Finally, I thank my wife, who has helped me beyond measure over the past two years, and who typed this thesis.

"But Scientists, who ought to know, Assure us that it must be so ...Oh I let us never, never doubt What nobody is sure about I"

Hilaire Belloc

Contents.

Page

MATERIALS 9

CHAPTER 1 INTRODUCTION. 10

CHAPTER 2 GAS-LIQUID CHROMATOGRAPHIC 66

ASSAY.

CHAPTER 3 TISSUE DISTRIBUTION AND 89EXCRETION.

CHAPTER 4 IN VITRO METABOLISM. 119

CHAPTER 5 PRELIMINARY TOXICITY 139EXPERIMENTS.

CHAPTER 6 INTERACTIONS WITH CYTOCHROME 159P-450 AND DRUG- METABOLISING ENZYMES.

CHAPTER 7 DISCUSSION. 224

REFERENCES 249

APPENDIX 268

Materials.Unless otherwise stated in the text, standard

laboratory chemicals were obtained from normal sources.Chlorofluoroalkanes were obtained from Cambrian

Chemicals Limited, Croydon, U.K. and British Drug Houses Chemicals Limited, Poole, U.K.

Enzymes and cofactors were obtained from the following sources: Sigma Chemical Company Limited,London, U.K. (glucose-6-phosphate, sodium salt; glucose- 6-phosphatejdehydrogenase ; NADP sodium salt; phospho- lipase C from Cl. welchii; phospholipase D from cabbage), Boehringer Corporation Limited, London, U.K. (glucose-6-phosphate, disodium salt; glucose-6-phosphate

+ \ dehydrogenase; NADP disodium salt), Whatman BiochemicalsLimited, Maidstone, Kent, U.K. (glucose-6-phosphatedehydrogenase) and International Enzymes Limited, London,U.K. (glucose-6-phosphate, disodium salt; NADP sodiumsalt).

CHAPTER 1

INTRODUCTION.

Contents.Page

1.1 Properties. 12

1.2 Uses. 16

1.3 Human exposure. 23

1.4 Toxicity - animal studies 24

1.5 Toxicity to man. 34

1*6 Anaesthetics. 36

1.7 Biological cleavage of the C-Cl bond. 39

1.8 Stability of the C-F bond in biologicalsystems. 47

1.9 Biological cleavage of the C-F bond. 48

1.10 Metabolism of the fluorinatedanaesthetics. 50

1.11 Metabolism of the chlorofluoroalkanes. 58

1.12 Biochemical aspects of foreign compoundmetabolism. 57

1.13 Effects of chlorofluoroalkanes andrelated compounds on drugmetabolism and cytochrome P-450. 6 l

1.14 Introduction to the present work. 64

1.1. Properties.The chlorofluoroalkanes are hydrocarbons,

substituted partially or completely with chlorine and fluorine. The discussion will be limited principally to the analogues of methane and ethane, as these are the most important commercially.

The chlorofluoroalkanes are all volatile compounds in comparison to the chlorinated analogues. Table1 . 1 shows the boiling points of fully substituted methane and ethane series. They are also very stable compounds chemically.

It can be seen from Table 1.2 that the Van der Waal's radius of the fluorine atom is considerably smaller than the radii of the other halogens and that it closely approaches the size of the hydrogen atom. Fluorine is also the most electronegative element known. These properties allow the fluorine atom to approach other atoms very closely resulting in stable bonds and high dissociation energies. A single C-F bond requires about 28 kJ/mol more to break it than does a C-H bond. The substitution of a fluorine atom for a chlorine atom would therefore be expected to modify the compound's properties. From Table 1.3 it will be noted that the bond energy term, which is derived from the compound's heat of formation, decreases slightly for the C-Cl bond with increasing fluorination.This would suggest that the C-Cl bond in dichlorodifluoro methane would be more easily broken than in carbon

Table 1 . 1

Boiling points (°C)

CH^ -164CF^ -129CCIF^ “8 l.lcci2f2 -29.9CCl^F 23.8CCl^ 76.8

C2H6 -88:6C2F6 “79C2C1F5 ■ - 3 8

CC1F2-CC1F2 3 . 8

CC10F-CC1F 47.6CC12F-CC12F 93

C2C1 5F 135

C Cl. 1882 o

Properties

of ha

loge

ns

/-“NH0a0 \

<H!a '—'bOU P0 Fid 00 rQp 0

rH0 bOCQ FI

•ri

Wca ■H VO ON VO

0 VO CA vo ON CA•p LA CA CM

orH CO VO O CA

0 CM rH•p CA CM CM

iou

-pooHm

-p•H>•H

-PcCb O0F*

O O CO ■H• • • • •

CA CM CM CM

OVOONrH

bOti•HHEJctfPhsO

H0cC£U 0 TJFlctf>

«<w•HPcC

LA O LA LA OCA CO ON rH CM

• • • t •rH rH rH CM vH

HO FiCQ H ffi

00 o 00 A -fa CM •H1 1 CA CA CA CA

* o • • • •* rH rH rH rH£

•pbOfl<1)HTJflOCQ

o<$

m0fl

Hctf>ObO

H LA o A - LA <tfO NO A - A - 00 a1 h- A - A - 1 0o • • • • >

rH rH , rH rH a

mA•pbOfla)H

f l fa0 -'-N fa NO NO f l

•H H 1 1 CM 1 1 o O-P O O ■ LA &f l a

•H \ faO fa • f l0 0 f lW w f lCQ H m

•H Ja f l aT5 bO > 0

CA f l a• TJ 0 rH + fa •p

rH f l f l o rH A - 00 0 1 CO rH 1 1 -P |A0 CQ O CM CA CA fl bO

rH H f l& fl 0fl O flfa H 0fl

ooLA

ONLA NO

fa ON CM ON LA ON ON NOi 1 NO A - A- CO rH rH ON

{A H o 'w ' s -/ rHbO 0 •>—'fl a H fl0 \ H 0 >Nfl fa 0 rH fl0 M fl " X 0

-P o bOfa -P 0 0fl a 0 H fl0 a * * O CO COCQ 0 H /-s

+> O A - LA rH LA O a a a1 CM CM CM CM » 0 0 0o CA CA CA CA CA a a fl>_' fa fa fa

nflo»Q

Ofl'flfl|Afl

-P(DBOCDbO

0flH

'doBfl■ w w fl

** * + + + **

+ + CMfaCA

faCM

CAfaH H H HO O O O faO O o o o

va

tetrachloride. However, experimentally determined C-Cl bond dissociation energies for carbon tetrachloride and chlorotrifluoromethane show that this bond is broken less easily in the fluorinated compound. No other experimental values are available, but Gregory (1 9 6 6) calculated a dissociation energy for the C-Cl bond in trichlorofluoromethane. This shows that approximately 10% more energy is required to break the C-Cl bond in trichlorofluoromethane than in carbon tetrachloride. Table 1.4 compares some of the properties of selected chlorofluoroalkanes and includes, for comparison, carbon tetrachloride and 2-bromo-2-chloro-1 ,1 ,1-trifluoroethane, the anaesthetic halothane. Substitution of a fluorine atom increases the volatility of the compound, and decreases the density, viscosity and surface tension. Water solubility is generally decreased, although trichlorofluoromethane is slightly more soluble in water at 30°C than is carbon tetrachloride (Table 1.4).

1.2. Uses.The synthesis of carbon tetrafluoride was first

achieved in 1926 by Leb eau and Damiens. Dichloro- difluoromethane rapidly followed (Midgley and Henne,1 9 3 0)? having been prepared specifically for use as a refrigerant, to replace the inflammable ethylene and the corrosive and toxic ammonia and sulphur dioxide. This compound showed the necessary temperature-pressure

Physical

properties

of some

chlo

rofl

uoro

alka

nes•

O0

<H OO cn /—

CA+> 1 ON VD

-P (C a IA•rH o • •W TJ • rH vHPi •H faO<D PiQ U 1

•rlH

OPi /"*N 0•H £\ in(M£+> •» CO rH•rl S o rHH •p • ••rl (ti o orQ u2 Q) rHrH •P0 (C -Pm £ ctf

'_/

(M on o oON in in VO(M on in COt • • • •rH rH rH rH rH

CO cn in03 in rH rHO ON O o on• • • • •o o o o o

^ So +>o <C CO co CO ON VO 00 CM« « • • • • •

rH VO on ON CO A- on O• A- 03 CM -tf4 in• A - CM OJft +> I• cOrQ

co ON ON ON• • • • • • •

cn o CM o A -in on CM O CO A- ON•H rH rH rH rH rH rH

CM CM .—sf t f t rH oH rH O PiO o Ph cfl

• O O CQ £!CM f t 1 1 S -P

f t ft CM ft CM o 0H f on CM H .CM ft 1 rH

H H H O rH rH on (3o o O ffi o O ft •flO O O O o o o '—'

relationships, together with the important uninflamra- ability and low inhalation toxicity. Chlorofluoro­alkanes are used for both domestic and industrial refrigeration.

The second principal use of chlorofluoroalkanes is in the field of aerosol propellants. The first application of fluorinated hydrocarbons to this use was in 1 9 3 3i but it was not until 1 9 ^ 7 that the present type of aerosol container, or can, was marketed (Pickthall, 1964). The use of liquified chlorofluoro­alkanes as aerosol propellants has advantages over the use of compressed gases, such as carbon dioxide, nitrous oxide and nitrogen. Firstly, compressed gas propellants must initially operate at very high pressures up to 90 p.s.i., and the containers must therefore be capable of withstanding such pressures with a consid­erable safety margin. The vapour pressure of a chlorofluoroalkane propellant mixture can be varied by altering the composition of the mixture. The vapour pressure of each compound is shown in Table 1.5* Secondly, as the product is dispensed from a compressed gas container, the gas must occupy a larger volume and the pressure falls. It is also possible to dispense propellant gas alone. With liquified propellants, however, as product is dispensed, some of the liquid propellant vaporises to fill the space, thus maintaining the internal pressure constant. Thus the propellant pressure depends on the vapour pressure of the components

Table 1.5 •

Vapour pressures of some chlorofluoroalkanes used asaerosol propellants.

b.p. (° C ) Vapour pressure (atm) at 21°C

c c i3f

CC12F2CC1F„c h c i2fCHC1F„CC1F -CC1Fd* d

2 3 . 8

•2 9 . 8

■8 1 . 4

8 . 9

•40.83.8

0 . 9 1

5.7832.21.579.341.88

as noted above. In addition, as product and liquified propellant leave the container, the propellant vaporises and disperses the product into fine particles. Thus the extensive use of chlorofluoroalkanes as aerosol propellants is due to a combination of desirable properties: the lack of inflammability, the apparently low toxicity and the fact that these compounds are volatile liquids or easily liquifiable gases.

The development of the market for pressurised aerosol dispensers has risen dramatically since 1947- In the United States of America sales are now well over 2000 million containers per annum. Progress in the United Kingdom has been rather slower. Table1 . 6 gives approximate sales figures up to 1 9 6l, together with figures of the production of chlorofluoroalkanes.

The aerosol packaging technique has been applied to a wide variety of products. Those for which the technique has some advantage are insecticides, room deodorants, hair lacquer and perfumes, paints and medicinal products. Some other applications are more for the novelty value and are generally more expensive and wasteful. These include cosmetics, personal deodorants (Goldberg and Netzbandt, 1 9 6 6 ), polish and some foods (Reed, 1 9 6 1). The medicinal products cover a wide spectrum. Aerosol skin sprays are used for applying dermatological products, for example in the treatment of athlete!s foot. Local anaesthetics are available in aerosol form; in fact

VO•

rH

0H&<c&

•prt VO(0 oH rHH0 Xft '—Ou TJft H0H w0W3 w0 -pJh •H0< p!

w IACO 03rH O

CMIACM

OIA

LALA

CMVO

< IA O O O O O CO• 1 11 1 LA A«- o O O r-

u.s IA CA LA VO CO orH

•<•W 100 t /—%

PI £> VO(S oM 0 rHH £(tf -P 140u PI U10 •H TJP* PiH 3<H 0 00 5h ftU P*0 -PrH o£ c3O <H2Picfl

a

VOONrH

0O>>ChCQ

a-o. u<H

O o O O O O O•sh A - O CM *=P O CMvH rH CM CM CM CA CA

VOONrH

HH£-P

o•HftsoJh

A - o LA VO A - CO ON O rH CMLA LA LA LA LA LA VO VO VO

ON ON ON ON ON ON ON ON ON ONrH rH rH rH rH rH rH rH rH rH

a pain-relieving sjDray is available that is purely a

mixture of trichlorofluoromethane and dichlorodifluoro- methane. It is intended for spraying on the source of pain or irritation in such conditions as shingles, and gives some relief by cooling the skin and reducing the sensory output of pain nerve endings. This latter effect may be due to either the cooling or to a mild local anaesthetic action.. Ellis (1 9 6 1) reported that chlorofluoroalkane propellant mixtures were more effective for the relief of muscular pain than ethyl chloride, which was probably the first aerosol produced, in a glass container, in l8 9 9 -

’Spray-on bandages' are chlorofluoroalkane propelled plastic film forming agents, and may be used pre- and post-operatively. They may be used on wounds and burns and have the advantage of painless application and transparency, so that the bandage may remain in place even for inspection of the area. Aerosol antiseptics are also used oh wounds and burns.

An even more intimate contact between user and aerosol spray is achieved with nasal, buccal and vaginal medications. Finally, drugs designed for inhalation may also be packed in aerosol form. These include drugs to counter respiratory infections and bronchospasm.

The les£ volatile chlorofluoroalkanes may be used on the industrial scale as specialised solvents. They are lipophilic compounds; 1 ,1 ,2 -trichloro-l,2 ,2 -trifluoro-

ethane, for example, is soluble in ethanol and is infinitely soluble in ether and benzene.

Related to their use as industrial refrigerants, chlorofluoroalkanes are being used to freeze food, particularly fish. Norwegian legislation permits residues of these compounds to a maximum of 2 0 0 p.p.m.

1.3 » Human exposure.Exposure of humans to the chlorofluoroalkanes

may be either accidental, incidental or deliberate, depending on the use. Accidental exposure is most likely to be a result of leakage of refrigerant from either an industrial or domestic appliance. At the other end of the scale, deliberate exposure may be occasioned by the use of pain-relieving sprays contain­ing only chlorofluoroalkanes (Ellis, 1 9 6 1), where the aerosol is directed on to the skin. Deliberate exposure has also occurred as an abuse (Bass, 1970); aerosol mixtures have been sprayed into plastic bags and the propellant gases inhaled. This is reported to give a pleasurable experience, and has been used as a substitute for glue solvent 1 sniffing*. Certainly, the compounds possess a pleasant, ether-like odour, but deeper subjective sensations have not been tested by the author. The majority of incidental exposures occur through the use of aerosol sprays in domestic and medical use. Popular use of room air fresheners, insecticides, paint, hair lacquers and perfumes has

greatly increased over the past few years. Medical uses of aerosol preparations have been described (1.2). There are many advantages for the presentation of some pharmaceutical products in this form. These include the obvious ease and convenience of application, together with speed. In addition, the product is kept free of contamination and remains potent for longer periods than usual, due ta the absence of air in the container. Further, by using a metered valve outlet, the dose can be accurately controlled. Thus the use of aerosols in the medical field would seem certain to increase.

1.4. Toxicity — animal studies.The biological effects of fluorine-containing

compounds have been the subject of several reviews.Hodge and Smith (196 5 ) adequately covered the biolog­ical properties of inorganic fluorides and the effects particularly on bone and teeth. Hodge et al. (1 9 6 3)presented, chiefly in tabular form, all the existing information on the fluorine-containing compounds then known.

Chenoweth and Hake (1 9 6 2) reviewed the anaesthetic and toxicological aspects of halogenated hydrocarbons and ethers containing up to four carbon atoms. Their review was not as complete for this group of compounds as that of Hodge et al. (1 9 6 3) as it was only intendedto cover the period 1955-1961. It is worthy of note

that only two papers are cited concerning work on the propellant chlorofluoroallcanes within this period.A recent and complete review by Larsen (1969) covers the field of clinically useful fluorinated anaesthetics.

Reviews of chlorofluoroalkane toxicity are provided by Clayton (1 9 6 2, 1 9 6 6 , 1 9 6 7a,b, 1 9 6 8).

As most of the chlorofluoroalkanes of interest are gases, exposure was most likely to occur by inhalation and early toxicological work was designed to evaluate the effects of short exposures, as might be experienced by householders and refrigeration engineers. The Underwriters Laboratories carried out simple acute exposures of guinea-pigs to various concentrations of the test compounds, and classified each compound on a scale running from 1-6 , where 1 indicates a highly toxic compound (causing death or serious injury when inhaled for 5 minutes at less than l%(v/v) in air) and 6 indicates a compound of low toxicity (causing no injury after a 2 hour exposure at 2 0%(v/v) in air). Table 1.7 shows this classification applied to three groups of halogenated methanes and ethanes, and also shows the Threshold Limit Values set by the American Conference of Industrial Hygienists in 1964. These values set the maximum permissible concentration in air, expressed as parts per million, on an eight hour average. It should be noted that several of the Threshold Limit Values for the chlorofluoroalkanes were set at 1000 p.p.m. not on the basis of toxicity,

of ch

loro

fluo

roal

kane

s

\ P

orH OoorH

ooovH

/-s /—\ /—s /—s /—\ ✓-N r—s ✓—\o o o O O O o o o aso o LA o o o o o o h-o o o o o o o o VOrH rH rH. rH rH rH rH rH as'_' '—' '— ' '_/ '_' v_/ rH

w

Pi0

-P

aS •H rHO <8

piu 0

LA as LA 0 •HVO VO CA 1 LA VO 1 VO •p V)

-cf -C}H <H •Has >

0* ufa

0U

-p pi•H toO 0

•H faX X0 06-h

rHas-pas

<h

toU0oA

+ +

(M 03 CM CM 03 CM • rH CM CM

CM O O O OrH 01 CM CM

CM O O OrH O] CM

CMfa fa ca

ca CM fa rH rH H H -ct’1 O O O O fa0 0 0 0 . 0

fa CM CA CM fa

H H HO O OW ffi Ko o oca

4 1 CO

LA

O O rH CM

fa CM CM CM fa fa

rH iH rH O O Oo.o-oI I Ifa fa CM

CM CM fa rH H rH O O O O O O

Pio•H-Paso•H<H•HCOCOCtf

HOCOCD

•Huo-puOrQR5

COPH0-p

•H

£Ph0TJPi

E>

.L.V.

- Threshold

Limit

Value

; values

in parentheses

are

but because adequate engineering standards should be capable of* adhering to this limit.

Other early work on toxicity was rather empirical and concerned with central nervous effects.

Sayers et al. (1930) continuously exposed dogsand monkeys to dichlorodifluoromethane and observed tremors and a staggering gait at '20% (v/v). They also found that growth was slightly inhibited in the first 2-3 weeks. Nuckolls (l933)» using guinea-pigs, found 3 - 1 0 % (v/v) trichlorofluoromethane for two hours produced loss of coordination, tremors, anaesthesia and convulsions; dichlorodifluoromethane was less effective. Cats tolerated up to 70% (v/v) dichloro­dif luoromethane in the atmosphere (Brenner, 1937)*Booth and Bixby (1932) investigated dichlorofluoro­methane and chlorodifluoromethane. Convulsions in mice occurred at k and l6% (v/v) respectively and death occurred at 20 and 49% (v/v).. Yant et al. (1932)studied 1 ,2-dichloro-l,1 ,2 ,2-tetrafluoroethane and found that approximately 2 0% (v/v) in air was required to kill dogs; guinea-pigs survived convulsions at this level of exposure. Henne showed in 1937 that chloror fluoromethane was rapidly lethal to guinea-pigs at lower atmospheric concentrations than dichlorofluoro­methane , a few per cent (v/v) compared to 10% (v/v) respectively.

It is apparent from Table 1.7 that the inclusion of one or more fluorine atoms in a molecule causes

a marked reduction in toxic effects. This trend is more consistent with fluorine substitution than with chlorine substitution; in the latter case increasing substitution first reduces toxicity (dichloromethane is less toxic than chloromethane), and then increases it (both carbon tetrachloride and chloroform are more toxic than dichloromethane).

The stabilising effect of fluorine on the toxicity of chloroalkanes may be due in part to the alteration of the physical and chemical properties of the molecule. Reference to Table 1.3 indicates that the substitution of one fluorine atom in carbon tetrachloride results in the fractional lengthening of the C-Cl bonds and a concomitant slight reduction in the bond energy terms. However, comparison of the bond dissociation energy values available for tetrafluoromethane, chloro- trifluoromethane and carbon tetrachloride indicates that for trichlorofluoromethane the energy required for a single dechlorination is probably a few kJ greater than for carbon tetrachloride.

Robbins (19^6) studied a number of fluorinated methanes and ethanes for their anaesthetic potency in mice. Many were discarded due to their provocation of convulsive responses, and the four compounds that appeared to be suitable were not investigated further.1 ,1 ,2-trichloro-l,2 ,2-trifluoroethane was examined by Carpenter ejt a]L. (19^9)- They found that death occurred in guinea-pigs exposed to atmospheres

containing more than 3 % o f the compound for 0 .5 - 1

hour.Lester and Greenberg (1950) included three chloro-

fluoroalkanes in their studies: trichlorofluoromethane, dichlorodifluoromethane and 1-chloro-l,1-difluoro- ethane. They found these compounds to be anaesthetic, and that no liver damage occurred by their inhalation. Acute exposures revealed trichlorofluoromethane as the most toxic of these compounds to rats, 10% being lethal within half an hour. In another publication they found that concentrations of 2 -3% of the two isomers of tetrachlorodifluoroethane were rapidly lethal to rats (Greenberg and Lester, 1950).

1 ,2-dichloro-l,1 ,2-trifluoroethane, 1 ,1 ,2 -trichloro1.2.2-trifluoroethane and 1,1,2,2-tetrachloro-1,2- difluoroethane were studied by Burn et al. (l959)»1 .2-dichloro-l,1 ,2-trifluoroethane was found to have similar pharmacological properties to halothane, but as it produced occasional cardiac arrhythmia, it could not be recommended for clinical use.

Chronic toxicological data has become more important since the introduction of chlorofluoroalkanes into fields other than mechanical refrigeration, particularly food propellants. Octafluorocyclobutanewas introduced as a food propellant in the United States in 1 9 6 1, following a 90-day inhalation study conducted by Clayton et al. (i9 6 0 ) using rats, rabbits,mice and dogs. This study detected no effect of

exposure to 10% octafluorocyclobutane for six hours daily on general appearance and behaviour, blood and urine parameters, or on the histopathological appearance of the organs. Chloropentafluoroethane was similarly studied with comparable results (Clayton et al., 1966). The chronic oral toxicity of dichlorodifluoromethane has been studied by Clayton (1 9 6 7a), dosing 1 6 0 - 3 7 9 mg/kg to rats and 8-4-95 mg/kg to dogs for a 90 day period. Slight changes in serum alkaline phosphatase and urinary fluoride levels were checked by a second experiment on rats at higher dose levels, equivalent to approximately 3 5 0 0 and 7000 p.p.m. of the diet. Urinary fluoride excretion remained within normal limits and slight changes in alkaline phosphatase were not directly attributable to dichlorodifluoromethane.

Scholz (1 9 6 2) reported acute and chronic toxicity studies of trichlorofluoromethane, dichlorodifluoro­methane , 1,1,2-trichloro-l,2,2-trifluoroethane, and1,2-dichloro-l,1,2,2-tetrafluoroethane, using rats, guinea-pigs, dogs and cats. No effect was found on blood, urine or liver. Skin and eye tolerance tests showed no adverse reactions.

The effects of inhalation of trichlorofluoro­methane and dichlorodifluoromethane on lung histology in laboratory animals have been reported by Kubler (1 9 6 3 ) 1 who found no changes, and Quevauviller et al. (1 9 6 3 ) 1 who reported some changes in the lungs of

mice exposed for 24 hours to 1% propellant vapour.Recently long-term continuous chronic toxicity

studies have been carried out on rats, guinea-pigs, dogs and monkeys with 1 % dichlorodifluoromethane (Prendergast et al., 1 9 6 7) and with 0.1% and 1% trichlorofluoromethane (Jenkins et al., 1970) toassess the possible dangers to humans living in a confined environment, such as nuclear submarines and space vehicles. In the former study, Prendergast et al. (1 9 6 7) demonstrated that continuous exposure todichlorodifluoromethane was more hazardous than repeated exposures for 30 days to approximately the same concentrations. However,, no significant effects of toxicity were found. Liver cell vacuolation in guinea-pigs and focal liver cell necrosis in monkeys .were detected histologically. Jenkins et al. (1970)found similar liver cell vacuolation in guinea-pigs after inhalation of 1000 p.p.m. trichlorofluoromethane for 90 days, and increased serum urea nitrogen in dogs. Clayton (1966) reported two experiments involving repeated exposures of rats to trichlorofluoromethane at 4000 p.p.m., 6 hours a day, for 28 days and at12,000 p.p.m., 4 hours a day, for 10 days. He found some histopathological changes including vacuolation of liver cells and brain neuroglial cells, oedema of the lungs and of brain neurones, and increased haemo­poiesis in the spleen.

The toxicology of some of the chlorofluoroalkanes

by oral and topical application has been investigated. Oral intubation of rats with 2 g/kg 1,2-dichloro-

1.1.2.2-tetrafluoroethane (Quevauviller, 1 9 6 5 )1 200-4.00 mg/kg dichlorodifluoromethane (Clayton, 1 9 6 6) and150 mg/kg chloropentafluoroethane (Clayton, 1967a)showed lack of toxicity by this route of administrationStudies in rats have shown 1,1,2-trichloro-1,2,2-tri-fluoroethane to be of low oral toxicity, and it isreportedly so in dogs and man (Clayton, 1967a).Michaelson and Huntsman (1964) showed the oral LD__50of 1, 1, 2-tri'chloro-1, 2 , 2-trif luor oethane in rats was43.0 + 4.8 g/kg.-

Instillations into the eye of a rabbit of 1,1,2- trichloro-1,2,2-trifluoroethane (Clayton, 1 9 6 7a) and dichlorodifluoromethane (Downing and Madinabeitia, i9 6 0 ) occasioned mild, transient corneal irritation.

Application to the skin, tongue, soft palate and auditory canal of the three pure compounds (trichloro- fluoromethane, dichlorodifluoromethane and 1,2-dichloro1.1.2.2-tetrafluoroethane) and two mixtures (trichloro- fluoromethane/dichlorodifluoromethane and trichloro-fluoromethane/chlorodifluoromethane) repeatedly for 6 weeks (Quevauviller et al., 1964) showed little effect except on the skin, which became oedematous and slightly inflamed. The reaction was most marked with 1,2-dichloro-,1,1,2,2-tetrafluoroethane and the trichlorofluoromethane/chlorodifluoromethane mixture and was more severe in older rats. Quevauviller et

al, (1964) also showed that burn and wound healing was retarded compared to controls.

Chlorinated alkanes are noted for their l)Lepato- toxicity e.g. carbon tetrachloride (Recknagel, 1 9 6 7 ) 1

chloroform, 1, 2-dichloroethane and 1, 1, 2 , 2-te.trachlor o- ethane (Irish, 1 9 6 2); the fluorine-containing halogenated alkanes on the other hand are not apparently hepatotoxic (Sayers et al., 1930; Nuckolls, 1933; Lester and Greenberg, 1950; Burn ert al., 1959; Clayton et al., I960; Clayton, 1966; Clayton, 1967a,b). Only mild liver cell vacuolation has been found after repeated exposures to chlorofluoroalkanes at levels that might be encountered by humans (Clayton, 1966; Prendergast et al., 1967;Jenkins jst aJL. , 1970). It is interesting to notethat studies by Nikitenko and Tolgskaya (1 9 6 5) on the toxicity of 1,2-dichloro-2-fluoroethane, 1-chloro-2,2-difluoroethane and 1,2,2-trifluoroethane to mice showed that the toxicity decreased with the substitution of fluorine for chlorine. They suggested that toxicity was dependent to some extent on the chlorine content.

Experiments performed by Epstein et al., (1967a,b)show that chlorofluoroalkanes may exhibit synergistic toxicity with other compounds foreign to the body. Trichlorofluoromethane, 1,1,2,2-tetrachloro-l,2-di-fluoroethane, 1,1,2-trichloro-l,2,2-trifluoroethaneand piperonyl butoxide (an insecticide synergist) were injected intraperitoneally into mice, either singly or in combination. After 51 weeks, the incidence

of hepatomata in male mice was found ±o be significantly increased in mice receiving mixtures of piperonyl butoxide and either ethane analogue compared to the single administrations or control animals. Epstein et al. (1 9 6 7a) indicated that unpublished data ofFujii and Epstein suggest that the chlorofluoroalkanes induce microsomal enzyme synthesis.

1.5. Toxicity to man.In 1968, Baselt and Cravey reported a fatal case

of inhalation of a mixture of trichlorofluoromethane and dichlorodifluoromethane by a fifteen year old boy, who had sprayed the contents of an aerosol can into a plastic b ag. They concluded that death was due to anoxia.

Bass (1970) reported a survey of sudden unexplain­able deaths in American youths, which were associated with inhalation of volatile hydrocarbon solvents, principally 1,1,1-trichloroethane and fluorinated refrigerants and propellants. These deaths were not due to suffocation by plastic bags, nor to freezing of the vocal chords and pulmonary oedema, but were associated in some way with stress or exercise. At post-mortem examination, eight cases showed levels of volatile hydrocarbons quantitatively compatible with light plane anaesthesia of similar anaesthetics. Bass did not reach any conclusion but suggested that severe cardiac arrhythmia, induced by the volatile compounds

and the stressful situations, may have been the cause of death. This report prompted Taylor and Harris (1970) to investigate the effect of chlorofluoroalkanes on asphyxia-induced cardiac arrhythmias in mice.They found that mice inhaling chlorofluoroalkanes, when compared to placebo and control animals, were more sensitive to arrhythmias induced by asphyxia, and the arrhythmias (sinus bradycardia, AV block and ventricular T-wave depression) proved fatal. They also reported preliminary findings that dogs and rats are also sensitised. The authors speculated that this ’sensitisation1 of the heart might be the cause of sudden deaths following chlorofluoroalkane inhalation, either as an abuse for pleasure, or by the use ofpressurised bronchodilator aerosols in asthma,particularly in circumstances favouring the endogenous release of catecholamines into the blood stream.This study by Taylor and Harris was criticised by Silverglade (l97l) who reported no differences between control and test animals in a similar study, and dismissed the postulated connection between asthma mortality and pressurised aerosol bronchodilators.Harris (l97l) briefly reviewed the possible causes of asthmatic deaths and concluded that pressurised broncho­dilator aerosols had little effect.

Reinhardt ejb aT. (197la) reported that high concentrations of the chlorofluoroalkanes, like other halogenated and non-halogenated hydrocarbons, do indeed

sensitise the heart to adrenaline resulting in severe arrhythmias. They concluded that sudden deaths following aerosol propellant sniffing may well be due

under proper conditions of use, aerosol propellants presented no risk with regard to cardiac sensitisation.

Blood concentrations of trichlorofluoromethane in human volunteers have been measured by Dollery et al. (1970) and Paterson et al. (l97l). Dollery et al.

this blood level by increasing the number of exposures. Paterson et al. confirmed these observations and reported Jack’s findings (1971) on the necessary blood level of trichlorofluoromethane to sensitise the hearts of conscious dogs to adrenaline. The peak blood levels shown by Dollery et al. and Paterson et al. were approximately 10% those of Jack.

Reinhardt et al. (1971b) exposed human volunteersto 0.1% and 0 .0 5% 1,1,2-trichloro-1,2,2-trifluoroethane repeatedly for a total of 30 hours. No adverse reactions were reported and analysis of breath samples indicated no significant build up of the compound in the body.

to cardiac sensitisation. They also noted that

showed levels of aboutof this chlorofluoroalkane. They could not increase

1.6. Anaesthetics.Halogenated hydrocarbons were first used as

anaesthetics with the introduction of chloroform in 1847-

Ethyl chloride followed in 1888 and trichloroethylene in 1935• Because of the hepatotoxicity of chloroform and trichloroethylene (Chenoweth and Hake, 1962; Irish, 1 9 6 2) and the explosion hazard with ethyl chloride and ether, fluorinated compounds were investigated for anaesthetic activity (Robbins, 1946). Although Robbins1 work was not followed up, fluroxene (2,2,2- trifluoroethyl vinyl ether) (Lu et al., 1953) 1 halothane (2-bromo-2-chloro-1 ,1 ,1-trifluoroethane) (Raventos, 1956)and methoxyflurane (2 ,2 -dichloro-1 ,1-difluoroethyl methyl ether) (Larsen, 1 9 6 3) have been widely accepted in clinical practice.

Halothane has caused considerable controversy since its introduction in 1956. Various reports have appeared of hepatic necrosis after halothane anaesthesia (Burnap ejt aT. , 1958; Stephan jert all., 1958, Virtue and Payne, 1958; Barton, 1959; Vourch et al., i9 6 0 , Brody and Sweet, 1963; Lindebaum and Leifer, 1963; Tygstrup, 1963; Keown and Bingham, 1969; Hansen, 1970; Nowill,1970; Johnston and Mendelsohn, 1971) • In 19631 the Committee on Anesthesia of the National Academy of Sciences-National Research Council in the U.S.A. set up the National Halothane Study. This was a retrospective study, concluded in 1966, of some 8 5 6 , 5 0 0

administrations of a general anaesthetic between 1959 and 1962. Of these, halothane accounted for 30%.The findings were that halothane appeared to exhibit a lower post-operative death rate compared to all

general anaesthetics. The subcommittee was unable either to confirm or to refute a causal relationship between halothane administration and the occasional massive hepatic necrosis. The incidence of hepatic necrosis in the study amounted to 82 cases, of which 9 were unexplained. Of these 9 cases, which resembled viral hepatitis, 7 had received halothane and 5 ofthese showed clinical evidence of liver failure. This contrasts with the explainable cases of necrosis, where liver failure was not clinically suspected. 2,3-Di- chloro-1,1,1,4,4,4-hexafluorobut-2-ene was found to be a contaminant of halothane (Cohen et al.-,' 1963) as manufactured by the high thermal process (Cohen et al., 1 9 6 5)- They further reported that it could concentrate under clinical conditions of use, that it was taken up by man and presumably metabolised, and that it was toxic to all species at levels only slightly above clinical concentrations. They reported that this contaminant was subsequently removed by the manufacturers

Some of the more recent reports of hepatitis (Johnston and Mendelsohn, 1971) and liver necrosis (Hansen, 1970; Nowill, 1970) associated with halothane anaesthesia have suggested a sensitivity reaction as being the cause; adverse reactions appeared only after two or more exposures to halothane. This is in agreement with the observations of Belfrage et al.\ (1 9 6 6 ) who challenged an anaesthetist, thought to be sensitive to halothane, with a few minutes halothane anaesthesia.

The subject fell ill within five hours, and although not jaundiced, the symptoms he displayed were attributed to the halothane challenge. The authors suggested a sensitivity reaction to some product of halothane, rather than the compound per se, as the effect was delayed.

Halothane is not the only fluorinated anaesthetic to be linked with hepatotoxicity. Methoxyflurane has been reported to have caused liver cell necrosis and jaundice.in one patient (Klein and Jeffries, 1966). Methoxyflurane has also been associated with nephro­toxicity. Panner et al. (1970) reported two deathsfollowing methoxyflurane anaesthesia, and gave evidence of crystals similar to calcium oxalate in the renal tubules. Taves et al. (1970) showed increases in the levels of inorganic fluoride and non-volatile organic fluoride in serum and urine following exposure to methoxyflurane. Fluoride excretion was prolonged,continuing for more than 19 days in one patient.Mazze et al. (l97l) reported several cases of nephro­toxicity.

1.7 « Biological cleavage of the C-Cl bond.The carbon-chlorine bond is cleaved in a number of

ways :-(i) Non-enzymic and enzymic reductive

dechlorination occurs in vitro, with a requirement for compounds such as glutathione or cysteine.

(ii) Dechlorination followed by mercapturic acid formation occurs principally with mono-substituted alkanes and various halogenated aromatic compounds.

(iii) Some oxidative processes occur in vivo resulting ultimately in the formation of carbon dioxide.

(iv) Homolytic cleavage of carbon-chlorine bonds occurs both in vivo and in vitro, resulting in the formation of free radicals.

Lucas (1928) studied the degradation of chloroform, tribromomethane and other brominated anaesthetic agents. He found tribromomethane was metabolised in vivo and in vitro, forming bromide ions. Chloroform, in in vitro incubations with rabbit or dog liver homogenate, formed chloride ions. In 1948, Heppel and Porterfield demonstrated the jdehalogenation by rat liver extracts of several partially halogenated methanes and ethanes.These included chloroform, bromochloromethane and dichloromethane and the products formed were formaldehyd halide ions and hydrogen ions. Sulphydryl compounds, such as glutathione and cysteine, activated the enzyme. Bray jet aJL. (1952) studied the dechlorination of twenty-nine aliphatic and aromatic chlorinated compounds including chloroform. They found that the liberation of chloride ions was increased in the presence of rodent liver extracts and cysteine, and that boiling the liver extract failed to reduce the activity markedly. They concluded that the dechlorination was

non-enzymic. Butler (l96l) found carbon tetrachloride and chloroform were metabolised to chloroform and dichloromethane respectively, by homogenates of mouse tissues. Heated tissue carried out this reaction more slowly.„ Chloroform was also converted to dichloro­methane by high concentrations of cysteine and ascorbic acid; carbon tetrachloride reacted more rapidly with these compounds and also with reduced glutathione and cytochrome c.

Booth jet al. (1 9 6 1) partially purified an enzymewhich formed glutathione conjugates from a number of halogenated compounds. Much work has been carried out on the formation of mercapturic acids from brominated alkanes, involving dehalogenation (e.g. Bray and James, 1 9 5 8 ; Thomson et al., 1963)* Barnsley (1 9 6 6) reported the formation of 2-hydroxypropylmercapturic acid from 1-chloropropane and l-chloropropan-2-ol. Grenby and Young (i9 6 0 ) had previously'■ detected n-propylmercapturic acid from the urine of rats and rabbits dosed with 1-chloropropane. Sims and Grover (1 9 6 5) found a close relationship between the production of chloride ions and the utilization of glutathione in the in vitro conjugation of 0C- 3 , 4 , 5 1 6-tetrachlorocyclohex- 1-ene and oc-2,3 1 4,516-pentachlorocyclohex-l-ene. Benzyl chloride is dehalogenated and conjugated (Boyland and Chasseaud, 1969)i whereas chlorobenzene is not dehalogenated, but is conjugated with glutathione at the 4-position

(Hutson, 1970)-

Another reaction dependent on reduced glutathione>

is the dehydrochlorination of DDT (1,1-bis(p-chlorophenyl) -2,2,2-trichloroethane) to.form DDE (1,1-bis(p-chloro- phenyl)- 2,2 -jdichloroethylene) (Hathway, 1970).

In experiments conducted by Paul and Rubinstein (1963)i chloroform and carbon tetrachloride were metabolised in vivo by the rat to 4% and 1 % carbon dioxide respectively, with approximately 8 0% of each compound being expired unchanged within l8 hours. In in vitro experiments with slices of liver, kidney, muscle and adipose tissue, small amounts of chloroform were found in incubations with carbon tetrachloride, but no dichloromethane was detected as a metabolite of either chloroform or carbon tetrachloride. Reduced glutathione and cysteine had no effect on the metabolic- . conversion. From these results, Paul and Rubinstein suggested that the formation of carbon dioxide from chloroform and carbon tetrachloride was by a common oxidative pathway, following the reductive dechlorination of carbon tetrachloride to chloroform. Van Dyke et al.

Ti4 1 ------(1964b) found approx. 3 % L Cj chloroform was excreted as "^CO and up to 2 % [36ci] chloroform appeared in the urine, principally as ^^Cl ion.

Trichloroethylene and tetrachloroethylene metabolism has been studied in the rat (Daniel, 19.63)*15% of the former and 2 % of the latter were excreted in the urine, principally as trichloroacetic acid and trichloroethanol. Daniel proposed the following

mechanisms to account for the intramolecular rearrangement of the chlorine atoms:

Cl ClW/ Y Cl Cl

Cl 0 Cl \ / \ / c — c / \ Cl Cl

Cl\Cl-C — C0C1 /Cl> CCl^COOH + Cl

Cl Cl\ = C / \ Cl H

Cl 0 Cl\ / \ /> c — c / \ Cl H

Cl\Cl-C/Cl

CHOCCl^COOH

CC1.CH OHj £

Jondorf et al. (1957) found hexachloroethane wasmetabolised by rabbits to a variety of products: trichloroacetic acid (l.3%)? trichloroethanol (1.320? dichloroacetic acid (0.8%), chloroacetic acid (0.7%)? dichloroethanol (0.4%) and oxalic acid (0 .1%) appeared in the urine; unchanged compound, carbon dioxide, tetrachloroethane and te tra chioro e thylene (12%) appeared in the expired air. Fowler (1969a) found only penta- chloroethane and tetrachloroethylene as metabolites of hexachloroethane in sheep. Yllner has investigated the in vivo metabolism of various radioactively labelled chlorinated ethanes in the mouse. His results can be summarised as follows:

SubstrateCH Cl-CH ClCHC12-CH2C1CC1 -CH Cl 3 2

CHC12-CHC12

Principal metabolites Referencec h2c i-coohCH C1-C00HCC1^-CH20HCCl^-COOH

C02CHC12-C00HCC12=CHC1*CC12=CC12*

Yllner (l971e) Yllner (l971d) Yllner (1971a)

Yllner (1971c)

Substrate Principal metabolites Referencec c i3-chci2 c c i2=chci Yllner (1971b)

CC12=cc12CC1 -CH(0H)2

The metabolites marked were formed spontaneously in neutral conditions (200 j x l substrate added to 100 ml 0.14M phosphate buffer, pH 7.0, containing 30% ethanol, incubated at 37°C for 24 hours).

Van Dyke and Wineman (l97l) have studied theo /T

dechlorination of Cl-labelled chlorinated ethanes and propanes in vitro, by measurement of the formation

O £of inorganic Cl ion. They found the activity was similar to the hepatic microsomal mixed function oxygenase (see 1.12.) in terms of cofactor requirements and inducibility by phenobarbital and benzpyrene.However, the dechlorination system appeared to be only slightly dependent on cytochrome P-450, as carbon monoxide failed to inhibit the reaction significantly. Glutathione was not a requirement for the dechlorination mechanism. Considerable differences were found in the amount of Cl ion formed enzymatically. 1,1-Di- chloroethane, 1,1,2-trichloroethane, 1,1,2,2-tetrachloro-ethane and hexachloroethane underwent considerable dechlorination; chloroethane, 1,2-dichloroethane, 1,1,1-trichloroethane and 1,1,1,2-tetrachloroethane were barely attacked.

Considerable attention has been focussed on the metabolism of carbon tetrachloride, which is now

thought to be associated with the compound's hepato- toxicity. McLean and McLean (196 5 , 1966) and Seawrightand McLean (1 9 6 7) showed that maintenance of rats on a protein-free diet, which depresses microsomal drug- metabolising activity, increased the oral LD-^ of carbon tetrachloride, decreased its metabolism to carbon dioxide by 7 0%, and ameliorated its effects on various parameters of liver damage. Pulmonary excretion of the compound was unchanged. They found that a single dose of DDT or three doses of sodium phenobarbital increased microsomal enzyme activity and carbon tetrachloride toxicity. They thus associated carbon tetrachloride toxicity with its metabolism, which seemed to occur in the endoplasmic reticulum.

McCollister et al. showed that monkeys absorbed about 30% of* an inhaled dose of [_ CJ carbon tetrachloride half of which was eliminated by the lungs over a 75 day period. A small quantity of C0Q was expired, and some non-volatile radioactivity was detected in the urine. Paul and Rubinstein (196 3 ) also showed metabolism to carbon dioxide in vivo, and these workers and Butler (1 9 6 1) showed the in vitro production of chloroform.

Reynolds (1 9 6 3) found that after a dose of [^cj- carbon tetrachloride, some of the activity was bound chemically to liver tissue constituents. Subcellular localisation of the radioactivity showed a 3•1 ratio between microsomes and mitochondria. This suggested

the formation of a reactive metabolite. Cessi et al. (1 9 6 6) in a similar experiment found radioactivity bound to peptides, possibly through lysine, and they believed that carbonyl chloride was the reactive metabolite of carbon tetrachloride.

Slater (1 9 6 6) speculated that carbon tetrachloride was metabolised to a free radical, perhaps *CC1 , by homolytic fission of a carbon-chlorine bond. However, Gregory (1 9 6 6) suggested metabolism of carbon tetra­chloride and chloroform may occur by electron transfer resulting in a free radical and a chloride ion. Wirtschafter and Cronyn (1964) had previously postulated a free radical mechanism as an explanation of halogen­ated and non-halogenated solvent toxicity. Both Slater; and Wirtschafter and Cronyn suggested the -participation of naturally occurring endogenous free

14 16radicals. Gordis (.1969) showed that when C- or Cl- labelled carbon tetrachloride was injected into rats, radioactivity was rapidly bound to a mixture of branched long-chain fatty acids; liver slices formed similar metabolites. Gordis also showed that chemically produced trichloromethyl radicals combined with methyl oleate, yielding esters with similar characteristics to the metabolites formed in vivo and in vitro.

Experiments by Fowler (1969b) on the metabolism of carbon tetrachloride by the rabbit showed chloroform (approximately 1%), hexachloroethane (less than 0 .0 0 1%) and two other unidentified metabolites. Dimerisation

of trichloromethyl radicals might have occurred to form hexachloroethane. Slater and Sawyer (1971a,b,c) have studied the increased lipid peroxidation due to carbon tetrachloride and conclude that it is dependent on the initial formation of trichloromethyl radicals.

Scholler (1970) suggests a similar metabolite in chloroform metabolism. Chemically, chloroform in the presence of free radical sources, e.g. benzoyl peroxide, yields the trichloromethyl radical (Walling, 1957)? and not the dichloromethyl radical as might be expected from comparison of the C-Cl and C-H bond energies in chloroform (Table 1.2). Chloroform and, to a lesser extent, carbon tetrachloride.form carbanions in the presence of a strongly basic nucleophile (Hine, 1950), but Slater (1972) discounted this as a possible mechanism of biological degradation, in view of the highly non-polar environment of the endoplasmic reticulum.

1.8. Stability of the C-F bond in biological systems.Because of the high bond energy of the C-F bond,

it is rarely broken biologically or chemically. The C-F bond in fluoroacetate withstands boiling concentrated sulphuric acid (Saunders and Stacey, 1948) and is only broken by sodium fusion at 500°C or by refluxing with 30% sodium hydroxide (Mirosevic-Sorgo and Saunders,1959). This property, together with the stabilising effect on other carbon-halogen bonds to the same carbon atom, has been put to use in the elucidation of some

biological phenomena. For example, cerebral blood1*131 1flow may be traced using [_ Al iodotrifluoromethane

(Solokoff, 1957; Kety, I960), whereas its analogue, iodomethane, is unstable in biological systems.

From the discovery that fluoroacetate was an inhibitor of the Krebs cycle by virtue of the formation of fluorocitrate (the classical example of 'lethal synthesis1.(Liebecq and Peters, 1949)) many fluorinated compounds have been assessed as selective inhibitors of various enzymes. 5-Flu.oro-2 *-deoxyur idine-5 1 -monophoshate, formed from 5“Pluorouracil, is a potent inhibitor of thymidylate synthetase (Hartmann and Heidelberger, 1 9 6 1) and thus blocks DNA synthesis.

Substitution of fluorine for hydrogen in molecules such as steroids has been used to stabilise the compounds and thus to decrease the metabolic degradation and consequent loss, of activity. Florini and Buyske (1959) showed that the metabolism of cortisol was decreased by 3 0% by the insertion of a fluorine atom in the 9C<- -position.

1.9. Biological cleavage of the C-F bond.Despite the great chemical stability of the C-F

bond, examples exist of its cleavage by biological systems. Goldman (1 9 6 9) reviewed the compounds containing the C-F bond which are of biological interest.

Two principal examples exist of C-F bond cleavage:(i) Peroxidase enzyme, derived from either

horseradish or turnip, was shown by Hughes and Saunders

(1954) to rupture the C-F bond of 4-fluoroaniline, yielding a fluoride ion. Similar reactions occurred with 4-chloro- and 4-iodo-aniline.

(ii) Phenylalanine hydroxylase obtained from rat or sheep liver was studied by Kaufman (l96l). 4-Fluorophenylalanine was found to be hydroxylated at approximately 1 5 % of the rate of phenylalanine hydroxylation. The reaction yielded L-tyrosine and fluoride ion in approximately equal amounts, utilising excess NADPH. 4-Chlorophenylalanine, however, forms3-chlorotyrosine by migration of the chlorine atom during hydroxylation (Guroff et al. , 1966).

Additional examples of C-F bond cleavage have been found by growing bacteria on^media enriched with various fluorinated carbon sources. Pseudomonads have been •found to defluorinate fluoroacetate, but not 2 - and 3 - fluor opr op1onat e (Goldman, 1965)? and 2-fluorobenzoic acid (Goldman et al., 1967) .

Barnett et al. (1967) described the defluorinationof oc-D-glucosyl fluoride by rat intestinal mucosa, and associated this activity with oi-D-glucosidase.

Daly et al. (1 9 6 8) reported the formation of4-hydroxyaniline from 4-fluoro- and 4-chloro-anilines, and the formation of the corresponding 4-hydroxy derivative from the 3~ and 4- position halogen substituted acetanilides, in a rat liver microsomal system.

Gottlieb et al. (1965) showed that 4-fluoro-L-

proline was incorporated into protein and collagen precursor in guinea-pig granuloma tissue, and was subsequently defluorinated to yield bound 4-hydroxy-L-

4-fluoro-L-proline.Recently, Ward and Huskisson (1972) showed the

defluorination of fluoroacetate by lettuce. The reaction was enzymic and required cysteine or reduced glutathione.

1.10. Metabolism of the fluorinated anaesthetics.Brown and Vandam (l97l) and Cohen (l97l) have

reviewed the metabolism of volatile anaesthetics, including the chlorinated and the non-halogenated compounds.

Until comparatively recently the inhalation anaesthetics were thought to be metabolically inert, with the principal exception of trichloroethylene. Vandam (1 9 6 3) suggested that halothane was 'inert1 because the carbon-chlorine and carbon-bromine bonds were stabilised by the adjacent trifluoromethyl,group. However, Stier (1964a) showed the prolonged urinary excretion of inorganic bromide by rats injected intra­muscularly with halothane in propylene glycol. The quantity recovered in the first three days represented 0 . 3 % of the dose of halothane administered. Using

average of 1 .5 6% of the injected dose in the 24-hour

proline. E. coli was also able to incorporate

Van Dyke et al. (1964a) found an

urine of* rats, which they assumed to be dechlorinated or debrominated residues, perhaps conjugated. They also found 0.84% of the radioactivity appeared as carbon dioxide within 24 hours. Stier (1964b) identified trifluoroacetic acid as a metabolite of halothane in rabbits. Van Dyke et al. (1964b), using• o/:Cl-labelled halothane, found inorganic chloride in rat urine for over l4 days following the dose. The amount totalled 2.9% of the dose given. 83- 90% of the radioactive material was recovered in the expired air within 30 hours. Van Dyke and Chenoweth (1 9 6 5b) found that the dechlorination of halothane also occurred in vitro with liver slices and liver microsomal preparations This dehalogenation showed requirements for oxygen and NADPH, but no requirement for sulphydryl compounds. Metabolism of halothane was also demonstrated in thymus and kidney slices. Van Dyke (1 9 6 6) extended this work to show that phenobarbital pretreatment of rats! increased metabolism of! [ * « ] halothane both in vivo and in vitro, by approximately 130% and 6 5% respectively.

Cohen and Hood (l9 69) and Cohen (1 9 6 9) have used low-temperature autoradiography to study the metabolism

T 14 1of [2- CJ halothane in the mouse. They found the highest concentrations of radioactivity immediately following inhalation to be in the liver and brown fat. Within ten minutes significant concentrations of non­volatile metabolites had built up in all tissues except the general body fat and brown fat. These metabolites

persisted in the body for at least 12 days and became concentrated in the liver and large intestine compared to the total body radioactivity. Cohen (1 9 6 9) showed that weekly injections of halothane caused a 400% increase in the_ non-volatile radioactivity of the liver over five weeks, and suggested halothane induced its own metabolism. The non-volatile metabolites in the mouse and squirrel monkey are mainly polypeptide and amino acid in nature, two-thirds having molecular - weights over 700 (Cohen, 1971 )• The liver metabolites appear to be closely associated with the mitochondrial and microsomal fractions. Sawyer et al. (1970), usingminiature swine, measured the extraction of halothane from the blood by the liver. Above 0.5% (v/v) halothane in the alveolar gas, extraction by the liver decreased •towards zero. They suggested therefore that the results of radioactive tracer studies of metabolism and uptake cannot necessarily be extrapolated to the situation of clinical anaesthesia and, conversely, that studies of metabolism under anaesthesia must grossly underestimate the degree of biotransformation at subanaesthetic levels. They found that the liver extracted nearly all the halothane from the blood, when the alveolar concentration was less than 0 .0 0 2 6% (v/v).

Sawyer et al. (1970) also reported blood levelsof halothane in clinical anaesthesiologists of between 0.6 and 3.5 yag/ml. Linde and Bruce (1 9 6 9) had reported

previously observations made of ten anaesthetists routinely exposed to 10 p.p.m. halothane in the operating room air. Halothane was found in the expired air of these subjects and the fluorine content of the urine was significantly increased.

Several studies have been made of the excretion of metabolites by patients following halothane anaesthesia (Stier et al., 1964; Rehder et al., 1967; Stier, 1968) and by volunteers following intravenous infusion of radioactive halothane (Cascorbi et al., 1971a,b) ortrifluoroethanol (Cascorbi and Blake, 1971)• Stier et al. (1964) found that patients anaesthetised withhalothane excreted bromide into the urine over a long period and that the bromideihalide ratio was increased. Rehder et al. (1967) found inorganic bromide andtrifluoroacetic acid in the urine of two patients,and calculated that between 12 and 20% of the halothane absorbed was metabolised. Stier (1 9 6 8) showed tri­fluoroacetic acid was excreted biphasically and that ~the time course was identical whether trifluoroacetic acid was given directly or was formed metabolically from halothane. He postulated the following mechanism for metabolism:

H F FI I I _C — B r ^F— C — C— Br fr-F— C— C00H + Br + ClI I I ICl F Cl F

as an alternative, or parallel, route to that proposed by Van Dyke and Chenoweth '(1 9 6 5a):

FIF — C IF .

F H FI I _ Ij> F C — C — OH + Br + Cl --->F— C— COOHI I IF H F

Trifluoroethanol has not been found as a urinarymetabolite, but Blake and Cascorbi (1970) using volunteers have shown up to 80% transformation of trifluoroethanol to trifluoroacetic acid. Cascorbi et al. (l971a,b) report large individual variationsin halothane metabolism. Comparison of a group of pharmacists and a group of anaesthetists showed that occupational exposure to halothane vapour can increase its metabolism. Comparison of identical and fraternal twins showed an overlying genetic influence on halothane metabolism.

It would seem from this evidence that halothane is debrominated and dechlorinated to form trifluoroacetic acid fairly readily, up to 25% of an anaesthetic dose being metabolised by man, and that defluorination accounts for a very small proportion of the metabolism (less than 1%). It seems possible that the route of metabolism might involve free radical formation, analogous to the metabolism of carbon tetrachloride (see 1 .7 ). The evidence of Cohen (l97l) for the formation of bound metabolites would seem to support this. Both Cascorbi et al. (1971b) and Cohen (1971) suggest that the combination of a free radical metabolite with a large protein might form an antigenic molecule. This could result in the 'sensitisation1 to halothane

F H I I F — C — C — Br I I F Cl

previously noted (1.6). Thus, the rare and unexpected cases of hepatotoxicity following halothane anaesthesia could be dependent on a number of factors, including the individual’s rate of halothane metabolism, the degree of enzyme induction produced by previous anaes­thesia or by other drugs, and the formation of antigens.

The metabolism of other fluorinated anaesthetics has not been as fully investigated. Fluroxene (2,2,2- trifluoroethyl vinyl ether) is not dehalogenated to any extent, but undergoes ether cleavage (Blake and Cascorbi, 1970; Cohen, 197l)s

CF<5CHq-0-CH=CH >CF CH_0H + H0CH=CH

CF^OOH C 0 2

Methoxyflurane is metabolised principally by dechlorination and ether cleavage (Van Dyke et al., 1964b; Van Dyke and Chenoweth, 1965a).

c h3-o-cf2chc i2 -— ^ c h3-o-cf2-c h2oh’ IHCHO + H0-CF - C H C 1 > H0CHoCHCl2 2 2 2

and induces its own dechlorination (Van Dyke, 1966). Defluorination also occurs (Holaday et al., 1970;Cohen, 1971) as indicated by the increased excretion of fluoride ion.

1 .1 1 . Metabolism of the chlorofluoroalkanes.Little, if any, study has been made of the possible

metabolism of the chlorofluoroalkanes. It wouldappear that the increased stability of the fluorine- containing compounds compared to the chlorinated analogues is assumed to confer metabolic inertness on the chlorofluoroalkanes. Saunders et al. (19^9)found l-chloro-2-fluoroethane non-toxic at a level of 0.184 mg/l of air, which was the for 2-fluoro-ethanol. l-Chloro-2-fluoroethane is unreactive to a variety of chemical reagents and the authors suggested that the animals were unable to metabolise the compound to form the toxic 2-fluoroethanol. Slater (1 9 6 5) compared the effects of trichlorofluoromethane (2.5 ml/kg) and carbon tetrachloride (l. 2 5 ml/kg) on serum ^.-glucuronidase activity and on liver levels of nucleotides in the rat. Whilst carbon tetrachloride produced a pronounced increase in the level of p- glucuronidase in the serum, trichlorofluoromethane caused a fall in the serum level of the enzyme, which was ’not appreciable’. Both carbon tetrachloride and trichlorofluoromethane caused a reduction in the ratio of NADPH to NADP in the liver compared to control levels in animals administered the liquid paraffin vehicle only. However, in the case of carbon tetra­chloride this was due to a decrease in the level of NADPH, whereas the level of NADP was increased by trichlorofluoromethane. From this evidence, Slater

concluded that metabolism of trichlorofluoromethane was probably insignificant.

1.12. Biochemical aspects of foreign compound metabolism.It is widely accepted that the vast majority of

compounds foreign to the mammalian system are metabolised usually to form less toxic metabolites, but occasionally to form an active, toxic metabolite (Parke, 1 9 6 8).The liver endoplasmic reticulum is a major site of this activity. Estabrook et al. (1963) showed that NADPH- dependent drug and steroid hydroxylations required the b type cytochrome, cytochrome P-450, as an oxygen activating system. The role of cytochrome P-450 in drug metabolism has been reviewed recently by Gillette et al. (1972). There have been numerous reviewarticles on other aspects of drug metabolism such as enzyme induction (Kuntzman, 1969; Gillette, 1971a), enzyme inhibition and activation (Anders, 1971) and the influence of other factors such as species and sex ' . .(Gillette, 1971b).

Cytochrome P-450 is found principally in liver microsomal and adrenal mitochondrial, preparations, although many other organs contain smaller concentrations Cytochrome P-450 is thought to be the terminal electron acceptor of the microsomal electron transport chain.A recent scheme, suggested by Estabrook et al. (l97l) for the microsomal electron transport reactions associated with cytochrome P-450 is shown in Figure 1.1.

The

microsomal

electron

transport

reactions

associated with

cyto

chro

me(from

Estabrook

et al

., 1971)

to

in-3-<-toift

oin

aoto

+++o o m o— -toift

++0 in

t o - < h :1&

i&

A

//

0wcO

S3 S30

•H feo0 0-P p0 a /Pif t r jO 0!> acO 1

H af t ft

Q<1a

aftQ§1

+++ i o Ioj in cm} ^--to--tf ;0

ft

in<P-CO o

M — --------

<DaOua m o a o-pIno

A

Q)sSi aI (D

CD! faO *P| o O 5.Pi 5ft9 a> CD cO t3 H | | <H HJla

\ <\ a A

aQ§

+> 1oS3O i U \ ft! T3 : ; cd ! -P : <0 ! : rH :: X01 a :! a

1aoto

wCDbOPiCO

aoCD>*H

-P•HmoftCD

a•pIn

a

a0-pco0•Hapi•Hw•HOin

1&0Sou

ao •O B -p o In -p o <00

a-p

fto 0a0 -pp>

co a- -p -p0-pcOu-pwaS3m

to

W *rl £0O TJ S3 CD 0 -P H cO cO *H > Oo0 m

a WfH cO

Of particular interest are the spectral properties

of the cytochrome. In chemically reduced (Na2* 2 4 conditions, CO ligands to the porphyrin prosthetic groups and causes a spectral change around 450 nin. By difference spectrophotometry this is shown as a sharp peak with a maximum at a wavelength of 450 nm and this is the basis of the assay of cytochrome P-450 (Omura and Sato, I964a,b).

\J

i--- 1--- 1--- » nm390 510

The second type of spectral interaction of particularimportance in drug metabolism is that produced by

vforeign compounds in aerobic microsomal preparations. Remmer et al. (1966, 1968) showed that addition of substrates such as hexobarbital, phenobarbital and aminopyrine to microsomal preparations gave a small difference spectrum (absorbance difference approx. 0 . 0 5

units) with a peak at approx 385 nm and a trough at approx. 420 nm. Substrates such as aniline and 2,4— dichloro-6-phenylphenoxyethylamine-HCl gave a difference spectrum with the peak at 430 nm and the

trough at 390 nm. These spectra are termed types I and II respectively:

II

350 390 430 470 350 390 . 430 470

Similar substrate-induced difference spectra have been obtained with human liver microsomal preparations (Kamataki et al., 1971)•

Microsomes contain relatively large amounts of lipid (approx. 370 ^g phospholipid/mg protein; Victoria and Barber, 1969)- It is now generally accepted that type I substrates are interacting with the phospholipid fraction of microsomes, as modification or extraction of the phospholipid by isooctane treatment (Leibman and Estabrook, 1971)? phospholipase C (Chaplin and Mannering,1970) and phospholipases C and D (Eling and DiAugustine,1971) destroys the type I spectral interaction. It has been suggested that type II substrates ligand to the haem of the cytochrome (Schenkman et al., 1967;Schenkman and Sato, 1 9 6 8).

The affinity of binding of substrates to microsomal, preparations can be estimated by measurement of the so-

called spectral constant (K ). Degkwitz et al. (1 9 6 9)S 1associated K with K in the metabolism of hexobarbital, s m ’but spectrally observable interaction (binding) does not appear to be an absolute requirement for metabolism as alteration of the type I site by isooctane or phospholipase treatment does not prevent metabolism of type I compounds (Chaplin and Mannering, 1970; Leibman and Estabrook, 1971)• .

1.13» Effects of chlorofluoroalkanes and related compounds on drug metabolism and cytochrome P-450.

Epstein et al. (1967a) referred to unpublishedobservations that chlorofluoroalkanes (trichlorofluoro­methane , 1 ,1 ,2 ,2-tetrachloro-l,2 7difluoroethane and1 ,1 ,2-trichloro-l,2 ,2-trifluoroethane) might be enzyme inducers, but this work appears not to have been followed up.

Van Dyke and Rikans (1970) showed that both —4halothane (8 x 10 M) and methoxyflurane (6 x 10 M)

enhanced the metabolism of aniline (a type II substrate) i-n vitro, but did not affect the metabolism of amino- pyrine (a type I substrate). Brown (1971) also studied the in vitro effects of halothane and found that type I substrate metabolism was depressed but that aniline 4-hydroxylation was enhanced. They also demonstrated that acetanilide 4-hydroxylation was actually depressed and zoxazolamine hydroxylation was unaffected. The lack of enhancement of the metabolism

of* these type II substrates is not in accord with the author1s suggestion that depression of.type I substrate metabolism causes increased metabolism of type II substrates by greater availability of electrons from the microsomal electron transport chain.

Davis et al. (l97l) found depression of type Isubstrate metabolism and cytochrome P-450 levels after injection of rats with halothane (lml/kg i.p.) for 2 days, but no effect on aniline 4-hydroxylation. However, the effects were smaller than those produced by carbon tetrachloride (0.4 ml/kg). Phenobarbital pretreatment increased the metabolic effects of carbon tetrachloride but not those of halothane. Berman and Bochantin (1970) showed that repeated exposure of rats to subanaesthetic concentrations (approx. 0 . 1 % (v/v) in air) of methoxy- flurane resulted in a non-specific stimulation of drug metabolism. Hexobarbital sleeping times were reduced markedly, aminopyrine demethylase activity increasedand the LD__ of methoxyflurane was increased.50Microsomal enzyme inhibitors reversed these effects.

The effects of chlorinated hydrocarbons, and particularly carbon tetrachloride, on drug metabolism have been widely studied. Dingell and Heimberg (1968) showed that carbon tetrachloride prolonged hexobarbital sleeping times in rats as early as 30 min after feeding orally 2.5 ml CCl^/kg body weight. Microsomal metabolism of hexobarbital, aminopyrine and 4 -nitrobenzoic acid was depressed by carbon tetrachloride administration.

In comparison, chloroform and dichloromethane caused only slight impairment of drug metabolism. Vorne and Alavaikko (l97l) showed increasing prolongation of hexobarbital sleeping times over a period of weeks by repeated doses of carbon tetrachloride. Lai et al.(1 9 7 0) showed similar effects due to inhalation of relatively low levels (about 3 00 p.p.m.) of carbon tetrachloride. Barker et al. (1969)? Feuer andGranda (1970) and Vorne and Arvela (l97l) have all reported depression of various drug-metabolising activities in rats by dosing the animals with carbon tetrachloride.

Glende (1972) has suggested that this depression ofI

drug metabolism is a direct result of lipid peroxidation damage of the endoplasmic reticulum. Wills (l97l) showed the inhibition of drug metabolism in vitro by enhancement of lipid peroxidation and believed that peroxidation controlled the integrity of the microsomal membrane and could thus exert a regulatory influence on drug metabolism.

Sasame et al. (1968) showed that carbon tetrachloride caused significant impairment of cytochrome P-450 in vivo in less than 3 hours. In in vitro incubations with microsomal preparations and NADPH, carbon tetra­chloride has been reported to convert cytochrome P-450 to cytochrome P-420 in a stoichiometric manner (Heni,1 9 7 1), with over 30% conversion occurring within 30 min.

Carbon tetrachloride has been shown to be a type I

compound (Reiner and Uehleke, 1971)* - However, of particular interest was their finding that Na^S^O^ abolished the type I spectrum and caused the formation of a difference spectrum very similar to the cytochrome P-450-C0 complex, with a peak at 454 nm. With microsomal preparations from phenobarbital pretreated rabbits, an absorbance change at 454 nm of approximately 50% that produced by CO at 450 nm was achieved. Reiner and Uehleke believed that this interaction was a liganding of the carbon tetrachloride directly to the haem of cytochrome P-450 and that this interaction was of fundamental importance in carbon tetrachloride toxicity.

1 .l4. Introduction to the present work.It has been shown that although there is a

considerable body of literature on the metabolism of chlorinated hydrocarbons and their effects on cytochrome P-450 and drug metabolism, little work has been reported on their analogues, the chlorofluoroalkanes.This has been due presumably to the low acute and chronic toxicity demonstrated for the chlorofluoroalkanes.

However, the existence of a few,reports of adverse effects, quoted in 1.4 and 1.5? indicates that chloro- fluoroalkanes are not biochemically inert. The fluorinated anaesthetics, although not all strictly chlorofluoroalkanes, have profound physiological effects. Evidence has been reviewed showing that they affect drug metabolism both in vitro and in vivo, the latter

effects being a possible factor in the occasional cases of toxicity to humans following anaesthesia with these agents.

The increasing use of chlorofluoroalkanes in applications that may result in the exposure of the general population implies that any toxic risk, if it exists, is becoming greater. In view of the profound toxic effects of the chlorinated hydrocarbons, which are thought to be due to more active, toxic metabolites formed by the mixed-function oxidase of the hepatic endoplasmic reticulum, it is desirable to assess whether metabolism of the chlorofluoroalkanes occurs, and to what extent under what conditions.

In our society at present, the use of sophisticated drug preparations and the ingestion or inhalation of a wide variety of other foreign compounds, such as food additives and dry-cleaning solvents, presents the danger of interactions between these compounds.Alteration in the normal biological fates of some foreign compounds can have unfortunate consequences, e.g. warfarin and phenylbutazone, ethanol and barbiturates. The effects of chlorofluoroalkanes on drug metabolism are therefore of importance as humans are frequently exposed to mixtures of these compounds with other foreign compounds.

The two basic areas covered by this thesis are the distribution and metabolism of chlorofluoroalkanes and the effects of these compounds in vivo and in vitro on cytochrome P-450 and drug metabolism.

CHAPTER 2

GAS-LIQUID CHROMATOGRAPHIC ASSAY.

Contents.Page

2.1 Introduction. 68

2.2 Apparatus and materials. 70

2.3 Column preparation. 71

2.4 Column efficiency. 72

2.5 Extraction solvent. ?4

2.6. Preparation of standard plots. 75

2.7- Separation of mixtures of chlorofluoro­alkanes. 85

2.8 Summary. 85

2.1. Introduction.Several techniques have been used to detect and

quantitate volatile chlorofluoroalkanes. Burns andSnow (1961), for example, estimated halogenated anaesthetics in body fluids by measuring the halide liberated by sodium reduction of the sample. Chenoweth et al. (1 9 6 2) used infra-red analysis for chloroformand methoxyflurane and X-ray fluorescence for halothane in monkey tissue samples. Silverman et al. (1 9 6 6)described infra-red spectrophotometric and gas chromato­graphic techniques for analysing aerosol propellants, but gave no data on quantitation. In the majority of investigations, however, gas-liquid chromatography has been the method of choice.

Quantitative determination of chlorofluoroalkanes •by gas-liquid chromatography has been applied principally to the quality control of aerosol products (Brook and Joyner, 1966; De Becker et al., 196?; Esposito and Swann, 1967; Konig, 1967; Bourne and Murphy, 1 9 6 9) and to the estimation of halothane in blood (Purchase, 1 9 6 3 ; Brachet- Liermain et al., 1971; Allott et al., 1971)- The measurement of chlorinated hydrocarbons in the expired air of humans has been carried out by Stewart et al.(1 9 6 3; 1 9 7 0a,b), using electron-capture detection gas-liquid chromatography. Boethner and Muranko (1969) used flame ionisation and thermoconductivity detectors for their work on halogenated hydrocarbons in the expired air of rats.

Extraction of chlorinated hydrocarbons from tissue samples into iso-octane was reported by Ehrner- Aamuel (1 9 6 3)- Butler and Hill had used n-heptane extraction (1 9 6 1). Carbon tetrachloride was used by Allott et al. (l97l) to extract halothane from tissues.Direct transfer of living tissue to the gas chromatograph using a special biopsy needle has been used by Cohen and Brewer (1964) for the analysis of halothane and 2,3“di- chloro-1,1 ,1 ,4,4,4-hexafluorobut-2-ene. The compoundsmeasured were all liquid at room temperature. The fluorinated analogues of interest in this study, however, are mostly gaseous (Table 1.4), which gives rise to additional problems.

The gas chromatographic conditions described in the literature for the separation of halogenated hydrocarbons are very diverse. The liquid phases range from silicone oils (Butler and Hill, 1961; Curry et al., 1962; Brook and Joyner, 1966; Ratcliffe and Targett, 1969; Allott et al., 1971) and Carbowax (Bourne and Murphy, 1 9 6 9) to more unusual mixtures, for example 3 % dibutyl phthalate and 'aluminosilicate1 catalyst (Petrova et al., 1968). Gadsen and McCord (1964) and Paulet et al. (1969) used 10% bis(2-ethylhexyl)sebacate and 2 % dioctyl sebacate respectively. Separations have also been achieved on porous polymer beads (Konig, 1 9 6 7 ; Foris and Lehman, 1969; Gvosdovich and Jashin, 1970) and grains of magnesium fluoride (Chichugova et_ al., 1964).The detergent fTide' has been used in the estimation of

halothane (Brachet-Liermain et al., 1971)• Otheroperating conditions varied as widely.

None of the methods reported seemed ideally suited to the present problem. As a result, an accurate quantitative assay by gas-liquid chromatography was developed which was applicable to solvent extracts of expired air, body fluids and tissues, containing trichlorofluoromethane or 1 ,2-dichloro-l,1 ,2 ,2 -tetra- fluoroethane. The electron-capture detector isparticularly sensitive to halogenated compounds (Clemons and Altshuller, 1966) and this detector was chosen for the assay.

2.2. Apparatus and materials.A Pye model 104 gas chromatograph, equipped with

6 ?a heated, 10 mC Ni electron-capture detector was used. Output was recorded, after amplification, on a Honeywell recorder. A Pye soap-bubble flow meter was used to measure the gas flow through the column.

Columns were made from 6 mm o.d. 3 mm i.d. annealed copper tubing, cleaned internally with several washes of acetone. Connectors supplied for use with glass columns were used to attach the metal columns to the detector inlet.

Samples were injected with a glass and stainless steel Terumo microlitre syringe.

Celite and analytical reagent grade solvents were obtained from BDH Chemicals, Poole, England. Squalane

and Apiezon L were obtained from Pye-Unicam, Cambridge, England. Trichlorofluoromethane, dichlorodifluoro­methane, dichlorof luoromethane and 1 ,2-dichloro-1 ,1 ,2 ,2- tetrafluoroethane were obtained from Cambrian Chemicals Ltd., Croydon, England.

2.3» Column preparation.A preliminary run with an existing column (l.5 m

3% oronite polybutene 128 and 0 .3% epikote 1001 on 80-100 mesh Diaport-S) showed that a packing ratio of stationary phase greater than 3%> together with a longer column, was required to separate trichloro- fluoromethane from diethyl ether. As a result, two different columns were prepared which proved suitable:

4.66 m 15% squalane on 80-120 mesh Celite2.0 m 15% Apiezon L grease on 80-120 mesh Celite.

An appropriate quantity (5 g per metre column length)was stirred with excess benzene in a 250 ml beaker andallowed to settle. The finer particles remaining in suspension were decanted off. This procedure was repeated three times. The stationary phase, squalane or Apiezon L grease (0.75g)? dissolved in benzene (15ml), was added to the Celite and the resulting slurry was well mixed. The benzene was evaporated off over a water bath, the beaker being shaken gently to ensure an even coating of stationary phase on the support material.The flocculent coated support was then dried, first in

AB

air and then under a vacuum over anhydrous silica gel.One end of the column tubing was plugged with

silica wool and the coated support poured into a funnel attached to the other end. Even packing was achieved by rapidly tapping the tubing on the floor. When the packing was a few centimetres from the top, a second silica wool plug was inserted. The tubing was then wound into a coil of 17-6 cm diameter. Columns were also packed with the tubing coiled. This required somewhat greater cafe in packing, but avoided to some extent Channelling1 of the packing material while coiling the tubing.

The columns were conditioned for 24 hours by passing oxygen-free nitrogen through the column with the oven temperature at the maximum level for the stationary phase (250°C for Apiezon L grease and 100°C for squalane). Table 2.1 shows the retention times of various volatile solvents on the two columns, under the conditions set out.

2.4. Column efficiency.The efficiency of a g.l.c. column, expressed as

theoretical plates, is given by the formula:

No. of plates = l6retention distance (cm)

peak width at half height (cm)

a good value lying between 300 and 3 0 0 0 theoretical plates/m.

Table 2.1.Separations obtained on column A.

Retention time (s) at column temperature of

Compoundair (injection artefact) CCl^FC H o C 1 9

(C2H5 )2°CHC1„n-hexaneCC1.

4o °c

924 4 l

447489

11731515229 8

6 0 C 92

3 26

326

4 n

756

Conditions - 40°C or 6 o °ccolumn temperature detector temperature 120°C carrier gas , flow rateelectron capture detector * checked with katharometer detector

40 ml/min

Separations obtained on column B .

Compound Retention time (s)34

101 110 153 259 2 8 0

484

Conditions - column temperature 6 l°Cdetector temperature 130°C other conditions as above

airCC13F(C2Ii5)2°CH2C12n-hexanec c i 4cyclohexane

Under the conditions given in Table 2.1, column A had an efficiency of 4l5 plates/m for diethyl ether, and column B 715 plates/m. The efficiency of column A measured with n-hexane was 1360 plates/m.

2.3« Extraction solvent.The solvent chosen for the extraction of blood

and tissue samples had to be lipophilic and of as low a volatility as possible, commensurate with a short retention time on the column. Diethyl ether was un­suitable because of its volatility. n-Hexane was used initially but was subsequently found to contain impurities. n-Heptane eluted immediately after carbon tetrachloride and was suitable in other respects.

It was essential that the solvent was stored away from possible exposure to chlorofluoroalkanes, as the use of an aerosol spray, in a fume cupboard some distance away, caused an appreciable contamination of some n-heptane left open to the atmosphere.

Considerable difficulty was encountered in the early part of this work in making up accurate standard solutions. This difficulty was eventually traced to the use of polypropylene stoppers, which dissolved appreciable amounts of the chlorofluoroalkanes, and on subsequent use, some of the compound was extracted.This problem was overcome by using glass apparatus throughout and by rinsing all apparatus before use with n-heptane.

2.6. Preparation of standard plots.With compounds as volatile as the selected chloro

fluoroalkanes, it was necessary to use an internal standard in the quantitative assay. The ratio of peak heights or peak areas of sample to standard are plotted against the mass ratio of sample to standard. Very accurate sample injection volumes are thus unnecessary.

Diehloromethane was too volatile and tailed badly on the column. Chloroform contained an impurity with a retention time very close to that of trichloro- fluoromethane, and was thus unsuitable. Carbon tetra chloride was chosen as the internal standard, although its retention time on the column was considerably greater than those of the chlorofluoroalkanes.

Because of the volatility of the chlorofluoro­alkanes and the low viscosity of these compounds and the solvent, all dilutions of standards or test sample were made by weight.

Figure 2.1 shows a plot of peak height against concentration of carbon tetrachloride. Initially,carbon tetrachloride was used at a concentration of

—6approximately 10 g/g solvent; as the sensitivity ofthe assay was increased, the concentration was reduced

—8to about 2 x 10 g/g solvent.It was found that the peaks for the chlorofluoro­

alkanes and for carbon tetrachloride were nearly symmetrical and that peak height and peak area ratios

Figure 2.1.peak height

(units)100 4-

80

60

40 ••

20

io“V 10“6 10-5concentration of CCl^ (g/g solvent)

were directly related (Figure 2 .2 ). Consequently peak height ratios were used to estimate the concentration of chlorofluoroalkanes.

Figure 2.3 shows the effect of different sample volumes on the plot of peak height ratio against mass ratio for trichlorofluoromethane and the carbon tetra­chloride standard. 0 . 3 and 1 jx,1 samples produced almost identical lines, whereas 2 p i samples produced a line of shallower gradient. A sample volume of approximately 1 l1 was used in all subsequent determin­ations .

Figure 2.4 shows the loss of linearity in the standard plot for 1 ,2-dichloro-1 ,1 ,2 ,2-tetrafluoro- ethane, following storage of the standard solutions. Because of this, standard solutions were made on the •day of use.

In addition, the sensitivity of the electron- capture detector varied slightly from day to day.Over thirteen standard plots for trichlorofluoromethane, the mean gradient and standard deviation was 2 . 3 6 +0 . 3 0 peak height ratio units/mass ratio unit.

The electron-capture detector showed a large difference in response to trichlorofluoromethane and1 .2-dichloro-l,1,2,2-tetrafluoroethane. The lower limits of quantitation were approximately 1 pg for trichlorofluoromethane and 100 pg for 1 ,2 -dichloro-1 .1 .2.2-tetrafluoroethane. The linearity of standard plots for the compounds was from 0 . 1 to 5 * 0 mass ratio

Peak height ratio

(CCl3F/CClZt)

Figure 2.2.

0.6

0.4

.2

0.1 0.2Peak area ratio (CCl^F/CCl^)

Figure

2.3-

Effect

of sample

size

on standard plot.

o

CM

O

0•H-P<su HOO-PAbO ft, •HoAMcS©ft

HooCM

•H

coo o

mass ratio

(CC1

0F/C

C1

rH OJ IN-

(d <d CdQ P Q< O ©

CM

CDu3bO•Hft

CDa

•rl-P£P•ri£POr-{ft

uCdTJ£cd

-PtoCDP<Ho*p•Hucd CDPi•HHftOto

<10'O'1-p

03VD

Pto•rlCD£

cd0ft

-=F H O O 0 \ H <t« P f t Cd CMPi H O

CM O

mass ratio

(C0C1

0F.,/CC1

units for the former and from 10 to 200 mass ratio units for the latter (Figure 2.5). In practice, samples were diluted to give peak height ratios close to unity and standard plots were prepared over much smaller ranges. Gas chromatographic conditions finally used for quantitative assays were as follows:

2 m 1 5 % Apiezon L column (column B) column temperature 100°C detector temperature 150°Ccarrier gas O^-free , flow rate 60 ml/minelectron-capture detector pulse space 150 yjusec

2amplifier attenuation 2 x 10 chart speed 40 in/hour.

Under these conditions, retention times in seconds were as follows:

air 25

o o H to to 30c c i3f 4 9 . 5

c h c i3 ■: 96CC14 138n-heptane 161

Total analysis time was about four minutes (Figure .2.6) It was found possible to overlap two determinations without affecting peak height ratios and without obscuring any peaks (Figure 2.7). This cut the time required for a duplicate determination from 8 minutes

Figure 2.6.

D

E

B

A injection point B airc ccif2 -ccif2

D CCl^F E CHCl^F CCl^G n-heptane

V/

Figure 2.7-

A injection point

B air C CCl^F D CCl^E n-heptane

DD

B

* indicates second run

B

E

V J \ J

E

to minutes.

2.7» Separation of mixtures of chlorofluoroalkanes.The conditions used for the quantitation of

individual chlorofluoroalkanes were not suitable for the analysis of mixtures. It proved impossible to separate dichlorodifluoromethane.and 1,2-dichloro-1,1,2,2-tetrafluoroethane on the 2 m column B (2.3)• However, separation was achieved on column A (2.3) of mixtures of dichlorodifluoromethane, dichlorofluoro- methane, trichlorofluoromethane and 1,2-dichloro-1,1,2,2- tetrafluoroethane. This is shown in Figure 2.8.

For metabolism studies, it was necessary to be able to detect dichlorofluoromethane in the presence of a large excess of trichlorofluoromethane (see 4.1) but in a short analysis time. Under the conditions given in Figure 2.9i using column A, it was possible to detect 0.21 j p g dichlorofluoromethane in the presence of 100 j x g trichlorofluoromethane per g solvent, in a total analysis time of 10 minutes. This time could not be reduced by heating the column to elute the trichlorofluoromethane more rapidly, as this destroyed the baseline stability.

2.8. Summary.A quantitative gas-liquid chromatographic assay

for solutions of trichlorofluoromethane and 1,2-di- chloro-1,1,2,2-tetrafluoroethane in n-heptane was

Figure 2.8.

C

A injection point B airC CC12F2 D CC1F2-CC1F2E CHC1 FF CC13F

D

E

Column A at 43°C and N2 flow rate 80 ml/min

Retention times (s)B 48 C 73 D 87 E 139 F 230

B A

Figure 2 .9.

Separation of 0.2% CHC1 ,F from CCl^F (100 p»g/ml)•

A injection point

BC CHC1 F£-kD CC10FE n-heptane

D C B

%r

Conditions - column A at 110°C, detector temperature 250°C, carrier gas flow rate 60 ml/min

developed, utilising internal standardisation with carbon tetrachloride. The difference found in the response of the electron-capture detector to these compounds agreed quantitatively with the findings of Clemons and Altshuller (1966). Qualitatively, the detector response to the chlorofluoroalkanes was CC13F> CCl^ » CF2C12 > CC1F2-CC1F2 » CHFC12, which again was in agreement with Clemons and Altshuller1s work.Two principal points emerged that required rigid . adherence. Firstly, glass apparatus had to be used throughout the handling of samples, and secondly, standards had to be made up by weight on the day of use.

A qualitative, and potentially quantitative, separation of mixtures of chlorofluoroalkanes was developed.

CHAPTER 3

TISSUE DISTRIBUTION AND EXCRETION.

Contents.

Introduction.

Experimental methods.Animals; extraction and assay of chlorofluoroalkanes in tissues;; extraction of chlorofluoroalkanes from the expired air.

Results.Recovery of chlorofluoroalkanes from tissues; accuracy of analysis tissue distribution pattern of 1,2 dichloro-1,1,2,2-tetrafluoroethane tissue distribution pattern of trichlorofluoromethane; trichloro- fluoromethane in the expired air.

Summary.

3.1. Introduction.Little study has been made of the uptake,

distribution and excretion of chlorofluoroalkanes.Paulet et al. (1 9 6 9) studied the retention of chloro-fluoroalkanes in the lungs of humans and dogs. They found that the concentration of these compounds in expired air fell exponentially, and that in the case of humans the chlorofluoroalkanes were expired within a few minutes. After the commencement of the work reported in this chapter, Dollery et al. (1970) showed trichloro- fluoromethane was absorbed into the blood after in­halation. Paterson et al. (l97l) extended Dollery1sobservations and estimated the blood half-life of trichlorofluoromethane in humans to be between 0.3 and 1.5 minutes.

3.2. Experimental methods.The animals used were male and female Wistar

albino rats (Porton strain, random bred in closed colony) weighing between l60 and 200 g.

Chlorofluoroalkanes were administered by stomach tube as solutions in ethyl oleate, at levels of 50 mS trichlorofluoromethane/kg body weight and 560 mg1,2-dichloro-1,1,2,2-tetrafluoroethane/kg body weight. Oral feeding was chosen for the following reasons:

(i) Large scale facilities for inhalation exposure were not available.

(ii) The dose given by stomach tube was

accurately known, as rats do not exhibit reverse peristalsis.

(iii) Oral feeding avoided the direct surfacecontact with organs that would have occurred if intra-peritoneal injection had been chosen.

Extraction and assay of chlorofluoroalkanes from tissuesExperiments to check the recovery of chlorofluoro­

alkanes in n-heptane in the presence of blood and tissues were carried out. Small weighed pieces of tissue from untreated animals were placed into known

Iweights of n-heptane, containing varying concentrations of chlorofluoroalkane. The tissue was homogenisedand the extract was assayed as described below.Standard solutions for these experiments were made up from dilutions of the n-heptane containing chlorofluoro­alkane . Recovery was calculated from the mass ratio of compound to standard found by assay compared to the mass ratio calculated for the original solution.

Extraction of chlorofluoroalkanes from rat tissues with n-heptane was effected as outlined in Figure 3«1«At known time intervals after dosing, the rats were killed by cervical dislocation or, if blood was required by guillotining. In the case of all tissues except blood, a piece weighing approximately 0.2g was rapidly removed and placed into a known weight of n-heptane (lOg). Samples of six tissues could be removed

Figure 3 .1.Extraction procedure for tissues.

Animal dosed orally

Killed after known time and tissue rapidly excised

Approx. 0.2g placed into weighed tube containing lOg n-heptane

Reweighed and homogenised with a'Polytron’

Approx. 0.05g extract into weighed volumetric flask containing 6 .8 3g n-heptane

(10ml)

Reweighed and CC1. internal standard added

Reweighed and mixed

10ml extract into volumetric flask

VWeighed and CCl^ added

VReweighed and mixed

Assayed by g.l.c.

within 60 seconds following death. The weight of tissue was determined by difference. The tissue was rapidly homogenised for 10-30 seconds with a Polytron PT10 ultrasonic homogeniser (obtained from Northern Media Supplies, Hull, England) and the tube re-stoppered. When the tissue debris had settled, 10 ml of the n-heptane was transferred to a glass-stoppered 10 ml volumetric flask. Carbon tetrachloride was added by weight on a balance accurate to four decimal places. In the case of.1,2-dichloro-l,1,2,2-tetrafluoroethane, this could be assayed directly by g.l.c. With trichlorofluoro­methane , however, a further 1:200 dilution of the n- heptane extract was required. In order to minimise possible variations in the concentration of the chloro- fluoroalkanes within each tissue, due to differences -in blood flow and lipid distribution, the pieces of tissue taken from different animals were removed from approximately the same areas:

liver - lower edge of right lobelung - part of inferior lobe of right

lungbrain - mid- and hind-braincardiac muscle - tip of ventriclesadipose tissue - epididymal fat pad in male, fat

posterior to right kidney in female.

Blood was treated differently. The rats were guillot­

ined and whole blood was collected directly into a beaker containing 15 ml n-heptane. Blood and solvent were rapidly poured into a 25 ml graduated glass tube and the volume of blood collected was noted. Extraction of the chlorofluoroalkanes was accomplished by shaking with the solvent, as sonication with the Polytron produced a highly viscous colloidal mass. The n- heptane extract was then assayed as described above.

Repeated blood samples from individual animals were taken from the tail vein. Samples taken by this method took between 45 seconds and 2 minutes to obtain.If the volume required (50 or 100 r1) was not collected within 2 minutes, clotting usually occurred within the pipette. The volume taken was then estimated by measuring the column of blood and subsequently weighing ~the same amount of water from that pipette. Blood samples were immediately blown out into 2 ml water, which was overlaid by 15 ml n-heptane, in a glass- stoppered tube. The pipette was rinsed out by sucking the water up and down it, followed by the n-heptane.The chlorofluoroalkane was extracted into the solvent by shaking the two phases together for about 60 seconds. The extract was then assayed without further dilution, as described above.

Extraction of chlorofluoroalkanes from the expired air.Trapping very volatile compounds in expired air,

particularly for kinetic studies, is difficult. Gage

(1 9 6 3) devised a complicated apparatus for extracting trichloroethylene from expired air with toluene.

However, it was decided that a simpler system would prove adequate.

The apparatus used is shown diagramatically in Figure 3*2. A fMetabowl! (Jencons Ltd., Hemel Hempstead, Herts, England) was connected to an extraction system containing n-heptane. Both sintered-disc bubblers and coil bubblers (Jencons Ltd.) were used.The former were simply glass tubes with sintered discs at the ends immersed in about 40 ml solvent. The principal disadvantage of this type was that the solvent was continuously bubbled through with air. Although chlorofluoroalkanes are very soluble in n-heptane, it was feared that the continual stream of air might remove the compounds from solution.In addition, over a period of hours, a considerable percentage of the solvent was lost by evaporation into the air stream. Loss of solvent by evaporation was small with the coil bubblers as the solvent (2 0 0 - 3 5 0 ml) was not continuously bubbled with air. Figure 3-2 only shows one of each type - in practice, six coil bubblers were used and five sintered-disc bubblers.All connecting pieces were glass and were blown to form butt joints, polyvinyl tubing merely being used to hold the joints close together. The only contact between the n-heptane and the polyvinyl tubing was at the bottom outlet; the tubing was clamped with a screw-

CM•

(TS0U0to•rift

m•pcoU<Hou•HcOT J0u•HftK00£-P

BoU<HCO0£COHCOouo3rHOUoHrCo<HO£o

♦rl-POCO-PK0uo<HW?!•P

ucOftft

-P -P

TO H 0 f i tU &

&

0•PrH•H<HO ■ •P

clip as closely as possible to the glass.The flow rate of air through the metabolism cage

was measured with a ’MeTeRate1 flowmeter (Glass Precision Engineering Ltd., Hemel Hempstead, Herts, England).The obvious position for this was on the air inlet of the fMetabowlf. However, when so placed, a resonant oscillation was set up in the meter. The manufacturers were unable to explain this and could not suggest a remedy. It was subsequently discovered that the oscillation was prevented by placing a buffer volume (an empty bubbler) between the meter and the VMetabowl1. This suggested that the volume of the metabolism cage was such as to permit the establishment of a standing wave pattern in the air flow. By placing the meter between the fMetabowlf and the first coil bubbler a slightly greater air flow was registered and the meter was thus placed in this position.

Air was drawn through the system by a filter pump at a rate of 200 ml/min.

Sampling the n-heptane in the coil bubblers was accomplished by removing the top caps, commencing at the farthest from the filter pump to prevent the solvent sucking back. At the end of each experiment, the volume of solvent in each bubbler was measured.Loss of solvent by evaporation was assumed to be linear and thus the solvent volume present at each sampling could be calculated.

Recovery of a known weight of chlorofluoroalkane

by the extraction system was tested by adding trichloro- fluoromethane to the first bubbler, drawing air through the system for 4 hours and assaying the amount of trichlorofluoromethane present in the entire system.

3.3. Results.Where appropriate, results have been expressed

as the mean + S.E.M. (standard error of the mean).* Student1s’ t-distribution was used to test significance Values of P of less than 0.05 were regarded as significant, and values between 0 . 0 5 and 0 . 1 0 as possibly significant.

Recovery of chlorofluoroalkanes from tissues.Table 3»1 shows the percentage of 1 ,2-dichloro-

1 ,1 ,2 ,2-tetrafluoroethane (concentration range 2 . 1 x 10

—6- 2 . 5 x 10 g/g n-heptane) recovered at different ratios of n-heptane to tissue (w/w). For lung,kidney, brain and cardiac muscle, a minimum ratio of approximately 1 5il was necessary for 1 0 0% recovery.For fat the ratio was in excess of 20:1 and for liver it was about 40:1. Brief studies with trichloro- fluoromethane showed similar results. It was decided to use a ratio of approximately 5 0 : 1 for tissue distrubution studies, i.e. 0 . 2 g tissue in 10 g n-heptane.

Recovery of chlorofluoroalkane from whole blood was lower. For example, using 15 ml n-heptane for the extraction of 4.5 ml blood, giving a n-heptane/

Table 3.1Recovery of* 1 , 2-dichloro-1, 1, 2 , 2-tetraf luoroethane from

tissue extracts.

Tissue n-heptane/tissue % recoveryratio (w/w)

Liver 2 88

k 9 k6 9 k

18 89k 2 100

Lung 6 627 87

1 2 . 5 9 8 - 1 0 0

Kidney 9 92-9511 98

Brain 13 9 k17 10221 93

Cardiac muscle 8 6 k15 100

28 101

3 k 9 k

Adipose tissue 7 739 71

12 80

21 97

50 101

blood ratio of 3 - 3 (v/v) , recovery was '5 l%i 5while 6 ml blood reduced the ratio to 2 . 3 (v/v) and the recovery to 3 6%. In view of the very low levels of chlorofluoroalkane subsequently found in the blood, and the inaccuracies inherent in the extraction of the blood, recovery was assumed to be 50% .

Accuracy of analysis.Duplicate g.l.c. analyses were made of each

sample. Peak height ratios for these duplicates were in good agreement (see Appendix Table A.l).

Dilution of the initial n-heptane extract and the addition of carbon tetrachloride as the g.l.c. internal standard were shown to be accurate to within about 5 % i by two parallel analyses of single tissue samples.For example, the following results for parallel analyses of liver extracts for trichlorofluoromethane showed good agreement:

(i) 1 2 . 0 and 1 2 . 5 j i - g / g liver(ii) 3 3 . 0 and 3 ^ . 0 j x g / g liver(iii) 5 3 - 0 and 5 6 . 0 j i g / g liver

However, analysis of two samples from a single tissue showed greater variation. In four separate experiments, the following concentrations of trichloro- fluoromethane were found in two samples of each liver:

(i) l6 . 0 and 6 . 0 j i g / g

(ii) 8.5 and 10.0 p g / g(iii) l4.0 and 11.5 f i g / g

( iv) 11.0 and 7-0 j X g / g .

Tissue distribution pattern of 1,2-dichloro-1,1,2,2- tetrafluoroethane.

Tissue levels of this compound were measured at various time intervals after an oral dose to male rats of 5 6 0 mg/kg body weight. Table 3 . 2 shows the mean concentration + S.E.M. of 1, 2-dichloroi-l, 1, 2 , 2-tetra- fluoroethane in various tissues at 0.5 ? 1 , 4, 6 and 19 hours. Figures 3-3 and 3-4 show the concentration of chlorofluoroalkane plotted against time for liver and adipose tissue and for lung respectively. Complete figures are given in the Appendix (Table A.2).

Maximum concentrations were reached at 0.5 hours in the lungs, at 1 hour in the blood and liver, and at4 hours in adipose tissue. Table 3-3 shows the ratio of the concentration of 1,2-dichloro-l,1,2,2-tetra- fluoroethane in the tissues to its concentration inthe blood. The liver maintained a concentration about 70 times that in the blood from 20 minutes to6 hours after dosing. Adipose tissue, however, showed a ratio rising to a maximum of 1100 at 4 hours. The ratio of concentration in the brain compared to blood reached only 1.5 and the value of 0.09 at 2 hours may have been primarily due to the chlorofluoroalkane in the blood flowing through the brain.

CM•

CA

0rH&CO

•P(0u0£•pd•H0dcO£

-P0ou0 drH<HcO?H-p0-p1CMCM

rHIOorH£0•rHTJ1CM

OCOH0>0

H

0WCO•rl£h

-P£b(•H0£>orQfcj[\bfsOVOLO

<HO0tooT5rH0odcOu0-pChcO

ON

<HO LAU0rQs§

OCA+ 1

LAO v-'LA

0-PcOo•rlTJCl•rl00to0£■pd0JhcOft

CO VDOCA + 1 wCOrH

d'rl00ShdfaO•Hft •+l H* I ’ [n-

co

Sfttm CO+ i

+ 1LA

CA0dtQto•d-p

TJ0COto0UftKft

2toto•H■P

<DCO

<Du3faO•H

rO

CO•H

VOCMO

2L

minu

tes

Figu

re

h-$— f* •

toHbO

CMH O O - O OO cn CM H

CM O

o• r l-P0UnooH&0

-P

CD2CQcq•H

■P

cfl

cqctf

TJCDCQCQCDShft«00S3Ctf43

ca -p• 0CA o

U0 oH 0,D HCC <HEh

Sh-P0

-P1CMe»03

ioShOH430

•H TJ1

03

rH

<HOPiO•H■PcCSh-PS30OS3oo

LA•

O O 03 IA oVD a - 03 1 03ca rH VD

O o OO 00 1 1 1 COca A-

OO o O ch

03 03 1 1 rH03 rH rH

O A - OsCO V0 rH 1 1 CArH '■. 03

•cq0

-pE*SJ Os

•H O LA Os 03 • • 1

rH LA Os o VDS3 VD LA•rl

0s

•rlEh LA rH

O • •VD 03 A - rH 1 A -

VD rH

LA CAO • •CO Os 03 rH 03 VD

A- 03 rH

-CAo •

03 CA A - 1 03 COIS rH

LA LAO • •rH rH A - 1 1

rH

0 0H 0O CQCQ CQ0 •HB -P

o 00 a CQ0 Eh S3 •H 0cq 0 bO •rl n ftcq > S3 a u •H

•H •H J3 u cC TJEh iJ CQ o <

Tissue distribution pattern of trichlorofluoromethane.Initially, very wide variations in tissue levels

were found, which were far greater than those experienced with 1 ,2 -dichloro-1 ,1 ,2 ,2-tetrafluoroethane (see Appendix Table A.3). Variations in the stomach content, affecting the rate of absorption of the compound, might have been a contributory factor and all subsequent experiments were carried out with animals starved for l8 hours.

Figure shows the blood levels of trichloro­fluoromethane found in male and female rats following an oral dose of 50 mg/kg body weight. At 15 minutes, the level was still rising in male animals. Females showed a blood level curve with two peaks. The concentration of trichlorofluoromethane at 7 minutes was higher than at 15 minutes (P<0.10) and significantly higher than at 30 minutes (P<0.05)« In addition, the concentration at 90 minutes was higher (P<0.10) than at 15 minutes. Differences between other time points were not significant.

Figure 3-6 shows the liver levels of trichloro­fluoromethane over a 90-minute period. In both male and female rats there was a sharp rise in the level within the first few minutes. In males, the liver concentration was not significantly decreased over the period from 5~30 minutes. However, in females, the initial rise was transient and a second peak occurred at about 40 minutes after dosing. The concentration

• CO

in

=§=*=»CMCO VO

Cn rQHO d) O H

uo>.

•riH

b O

i—IOobOJL

o OCM

OrH

O'CO

oVD

O

oCM

U<D>•rlHbO

rHO01

OCO

oVD

O

OCM

O OCM

OrH

//

minutes

minu

tes

of trichlorofluoromethane at both 7 and 45 minutes was significantly greater than at 20 minutes (P<0.05)«

Figure 3-7 shows the results for adipose tissue. The two-fold difference in concentration between male and female rats was almost certainly due to differences in blood supply between the epididymal fat pad and the supra-renal fat respectively. In the male, the maximum concentration of chlorofluoroalkane occurred at approximately 2 hours after dosing.

Complete results for blood,liver and adipose tissue levels of trichlorofluoromethane are given in the Appendix (Table A.4).

The double peak in the blood level curve (Figure 3 -5 ) was further investigated by analysis of serial blood samples taken from 6 female Wistar albino rats (starved for l8 hours). The individual blood level curves obtained are shown in Figure 3*8. It will be noted that five of the six curves showed two peaks and that, in two of the curves, the second peak was the larger. Figure 3-9 shows the result of averaging the curves over three minute intervals. Trichloro­fluoromethane reached a peak blood level within about 30 minutes and this level was not significantly lowered for a further 70-90 minutes (Appendix Table A.5)«

Trichlorofluoromethane in the expired air.

In three experiments, after an oral dose of 50

mg trichlorofluoromethane/kg body weight, 9 2 . 1 + 2 .6 . %

0iH(CSctfS0<H

0U2bO•Hfa

* * (M

VD

02toto•H•p0tooft•rlTJbO\fa<r\rHOOtodL

to0•pa•rlB

o o OCM

O■H

CO

0U2b O•Hfa

'dooH,Q<dHO££

Hs\facnHOObO

CO

CO

CO VO

'O

min

ute

s

On

03to

• r l

T5OOH,Q0rHOfH6\foHOO£

CO

CO

VD

minu

tes

of the dose was recovered unchanged from the expired air within 6 hours. Table 3*4 and Figure 3*10 show the results of these experiments.

Recovery of trichlorofluoromethane added to the first coil bubbler of the extraction system was 99 • 2% after 4 hours. The distribution of trichlorofluoro­methane between the six coil bubblers at this time was very similar to the distribution at 6 hours in the animal experiments (Table 3«5>).

3.4. Summary.A solvent to tissue ratio of 50:1 was required for

100% extraction of chlorofluoroalkanes from rat tissues with n-heptane.

Tissue distribution patterns were obtained for trichlorofluoromethane and 1,2-dichloro-1,1,2,2-tetra- fluoromethane after gastric intubation of rats with these compounds dissolved in ethyl oleate. Maximum tissue levels were reached at the times (min) shown below:

Dose Sex Blood Liver Lung Brain Cardiacmuscle

Adiposetissue

c c i3f M 15 + 5 - - - -

(5 0 mg/kg) F 7 & 60 7 & 45 - - - 90 +cc i f2-cc i f2 M 60 60 30 60 30 + 240

(5 6 0 mg/kg)

Assuming 4g liver, 1.4g lung and lg brain per lOOg body

Recovery

of trichloroflu

oromethane

from

the

expired

air

of the

female Wistar

albino

rat

•/—\

bOM\bOs

OLAv_'

0too*0HCtS

-p0•P

t 00 £s -P

•H•P Ch0-Pto 0Pi bO

•HcC •Pbf. Pi0

OU0f t

toa

*00toto0f t

LA LA• • •VO ON ON A-

CO co ON

ON CO• •

LA 1 CO rHCO ON

rH rH• •

-cF 1 VOA- A-

toP)3 LA O O0 • • •£ CA A~ CA .CA

VO VD LAs•H0s•H tH rH

• •CM 1 CO CA

CA

CA o• • •

•H VO CM CA•H t H rH

LA CO CA• 1 • •

O CM LA

bO

-PAbO•rl LA LA LA0 O O CM£ CM CM CM

Hcti£•HPi

<

-PPi0£•rlu0 0 © ©f t

Figure 3 • 10.Unchanged trichlorofluoromethane in the expired air.

Percentage of dose

100

80

6o

20

6k2

hours

Percentage

of the

total

tric

hlorof

luor

omet

hane

recovered

present

in each

coil

bubb

ler

VD03 03

O

in into

cn ino

-pPiCDa•Hu0ftX00A-pftoT )g00A-p-pGj

u0HrQ&3CQ

cn

03

COrH

oft

CO

CO

ftinft

COcn

*so 03• •r- VO

in ON VD• « • •■“ in. in

O n f t ON 03

Ol O• •

VO VOIN. VO

cor-ON

•PPi0S•HPi0ftX0>Nu0>oo0ft

wPiEJ0£

© stoPi

© /—\wPi

■p 3 •P 2Pi O PI 00 £ 0 £a s

•H VD •H VOPi v_'0 0f t f tX X

f t f t

weight (see Tables 3-3 and 3-5) and that approximately 1 0 % of the body weight of* rats (2 0 0g) was fat (Wardlaw et al., 1 9 6 9)? the following table shows the approximatetotal percentages of the dose of chlorofluoroalkanes that could be accounted for by these tissues at various times:

Time in minutes.Compound Sex 3 7 30 60 90 120 2 k 0

CCl^F M to • 00 1 1 - - .4-5 -CCl^F F 1 to • VJ1 1 2.9 .3-5 — -CC1F -CC1F

tL &M 1 . 0 1.5

Recovery of trichlorofluoromethane added to the extraction system was 9 9 •2 % after 4 hours operation.

In the female Wistar albino rat, 92.1 + 2 . 6 % of a dose of trichlorofluoromethane (3 0 mg/kg body weight) was expired unchanged within 6 hours.

CHAPTER k

VITRO METABOLISM.

Contents.

4.1 Introduction.

4.2 Methods.Animals; preparation of microsomal suspensions; in vitro incubation system; lipid peroxidation assay.

4.3 Results.In vitro incubations; lipid peroxidation.

Page121

122

130

4.4 Summary. 137

4.1. Introduction.It was possible that metabolism of trichloro­

fluoromethane might occur in a manner analogous to that proposed for carbon tetrachloride and chloroform. The following scheme was suggested:

method described earlier (2 .7 ) to detect 0 .2% dichloro- fluoromethane (I) in the presence of large concentrations

to detect the percentage of 1 ,1 ,2 ,2-tetrachloro-1 ,2-di- fluoroethane (II) that might- be expected to be formed (less than 0 .0 0 1%) under the g.l.c. conditions described.

The indirect method was to look for evidence of the formation of dichlorofluoromethyl radicals.Slater and Sawyer (1971a,b,c) showed that lipid per­oxidation was increased by carbon tetrachloride and, from a complicated series of experiments, concluded that this was due to the formation of trichloromethyl radicals. The production of a coloured complex by reaction of the malonaldehyde formed from peroxidised

X ? CHC12F (I)* CC12F-CC12F (II)

Two approaches were used to test this scheme, one

direct and one indirect. The direct approach was tolook for the possible metabolites in an in vitromicrosomal system. It was possible using the g.l.c.

of trichlorofluoromethane. It was i'easible, therefore,to investigate the production of (i) in vitro as apossible metabolite. However, it was not practicable

lipids with thiobarbituric acid (TBA) has been used to quantitate in vitro peroxidation (Ghoshal and Recknagel, 1965; Slater and Sawyer, 1971a). Thus lipid per­oxidation was used as a tool in attempts to detect radical formation from trichlorofluoromethane.

4.2. Methods.

Animals.The following species were supplied by the -

University of Surrey Animal Unit:

Wistar albino rats (Porton strain, random bred in closed colony), male and female in three weight ranges (6 0 - 7 0 g, 1 3 0 -1 5 0 g and 140-155 g)-

C^H/Mg strain mice (l8-22 g).

Syrian hamsters (golden - pure line strain) in two weight ranges (7 5 - 9 0 g and 140-155 g)*

Guinea-pigs (approx. 400 g).

Female chickens (5 months old) were obtained from a local farm.

Rats, mice, hamsters and guinea-pigs were housed in polypropylene cages on Sterilit bedding (W.P. Usher & Co. Ltd., London, England) and were fed on Spillers No. 1 Diet (autoclaved) (Spillers Ltd., Barking, Essex, England). The temperature and relative, humidity of the experimental rooms were maintained at 22°C and 50%

respectively. The lighting cycle was O6 3O-1 8 3O h (lights on).

Preparation of microsomal suspensions.Animals were killed by cervical dislocation and

the livers rapidly removed and cooled in ice-cold 1 .15% KC1. All subsequent manipulations were accomplished

I 0at 4 C. The liver was washed clear of blood, blotted dry, weighed and homogenised in 3 volumes of 1.15% KC1 .with a Potter-Elvehjem glass and teflon homogeniser, driven at 2950 rev./min. The suspension was centri­fuged at 10,000 g for 20 minutes at 4°C (High Speedav.l8 centrifuge with an 8 x 50 ml rotor, MSE Ltd., Crawley, Sussex, England). The supernatant was decanted and

■ Orecentrifuged at 4 C in a Superspeed 50 centrifuge(MSE Ltd.) either at 104,000 g for 1 hour (8 x 25 mlav.rotor) or at 1 5 7 ,0 0 0 g for 40 minutes (1 0 x 10 mlav.rotor). The supernatant (cell supernatant) was retained and for the lipid peroxidation assay the microsomal pellet was resuspended in sufficient 1 .15%KC1 to make 1 ml suspension equivalent to 1 g wet weight of liver. For the dechlorination system, the microsomal pellet was resuspended in 0.05M tris-HCl buffer pH 7-4 at a microsomal protein concentration of 5 mg/ml, measured by the method of Lowry et al.« (l95l)«

In vitro incubation system.A modification of the incubation conditions used

by Van Dyke and Wineman (l97l) in their radioactive assay of the dechlorination of ethanes and propanes was adopted. The incubation was carried out in 10 ml glass-stoppered tubes and the composition of the medium was as follows:

10 mg microsomal protein in 2 ml 0.05M tris-HCl buffer pH 7-4 1 ml cell supernatant 0.3mM NADP+

' 1 .5-mM glueose-6 -phosphate1 unit glucose-6-phosphate dehydrogenase.

Final volume 3*0 ml.

1 ^ and 10 jxl samples of trichlorofluoromethane were added to the incubation medium by microlitre syringe, immediately after flushing the tubes with oxygen. Incubations were for 1 hour at 37°C and at a shaking rate of 100 cycles/min. The reaction was not stopped by protein precipitation, but was stopped and extracted with 5 ml n-heptane. The n-heptane extract was assayed by g.l.c. (2 .7 ).

The system was later modified to ensure that no dichlorofluoromethane formed would be lost. The incubation was carried out in a 50 ml Quickfit round- bottomed flask, fitted with a separating funnel as a stopper. At the end of the incubation period, the

flask was cooled rapidly in ice and 5 ml n-heptane placed in the separating funnel. When the tap was

opened, the solvent was drawn into the incubation flask.The apparatus was shaken for a few minutes to extract the chlorofluoroalkane from the incubation medium, and the extract assayed by g.l.c.

Lipid peroxidation assay.Lipid peroxidation is expressed in terms of the

formation of malonaldehyde J (malonyldialdehyde, malonic dialdehyde). However, the thiobarbituric acid reaction is not highly specific (Franz and Cole, 1 9 6 2) and Saslaw et al. (1 9 6 6) have shown that the oxidation products of linolenic and arachidonic acids do not include malonaldehyd but other carbonyl compounds. The production of malon­aldehyde only represents a fraction of the destruction of fatty acids (May and McCay, 1 9 6 8) and is proportional to the amount of polyunsaturated fatty acids present (Bloom and Westerfeld, 1971)• Thus, the term 'malonaldehyde* is used here as a synonym for 'thiobarbituric acid-positive material'

The assay methods used by Ghoshal and Recknagel (19.65) and Slater and Sawyer (197la) are outlined in Figure 4.1 together with the modifications adopted.

Tetraethoxypropane (TEP), which hydrolyses to malon­aldehyde in acid solution, was used for the preparation of standard plots. Ghoshal and Recknagel claimed linearity of their standard plot from 0.44-4.4 jjig TEP/ml, but Scheig and Klatskin (1 9 6 9) only showed linearity from 0.l8 to 0.88 j Kg TEP/ml by the same method. Using Ghoshal and Recknagel1

In vitro

lipid

peroxidation

assay

meth

ods

0o oft -rl

HHH

f t •» Pi<ll f t

3 o O•H f t £3 70

3 0IQ O 0

/—•. 3 0 o +>H 0 f t tQ /--\H f t f t H'_/ f t £ H 70 H H .

0 O £ Pi '_■>0] ,Q IA LA 0

<n 3 IQ tQ 0O 70 l, so* os < < tQ3 0 r»- 0

•H 3 70 • Pl0 tQ O VO 0

H f t 2 f t b0£ f t tQ f t 0 0IA 0 (S 0 > u

• -P f t -P f t 70CM IQ <H tQ bQ 22070

HH

LA - -p• -3 LA 0CA 0 - <C 2CO ft CM 0 CQ ft t

70 ft • 70 f t + r- ft tQ TOPl Pl 0 O *rl H • 0 ft0 ft & g 0 •d vo 2 o/-V «• CO 0 t'- ft c0Pl 0 tQ ft -p x 3 X <D VO 10 ft h -h i 0 m •H ft > • VO o-p £ Pl VO O 70 £ ft o 1 f tcO (A \ ft 1 0 1 Pl b0 0 ✓—•. 0 Pift ft bQ CO ft 0 70 O ft H tQ Pl'_' a s 2 i •H tQ 3 0 ft £ >_/ 0 ft

cm £ 2 £ vo ■P 2 0 -P CM o f tf t Pl s—' CM £ O 1 0 tQ Pl CQ 2 20 0 • LA f t CO ,0 cO 70 < ft + <U ft 3b*, tQ ts- • H Pl H 0 O bo c070 0 CA LA ^ tQ £ O ft > f t 70 0 20 0 s + *p Pl 0 0 > • • O2 in 0 - »■ ft ft -ft ft b0 £ o 0

2ft

-p CQ ft O Q Pl • Pi 0 O -P 2 20 0 O • <1 3 VO H pl 3 f t W 0 70 f t2 3 2 CO 2 CA £ 2 10 -ft•H 2 2 2 • Pl LA Pi• c0 ft ft 70 £ £ Pl ft£ ft1

• •

a s ft s co -h CM ft CM 0 CM VO <1CO CQft

H

2 X 0CM b0 1 x 3 ft 70ft X 3 0 f t 3 •H 0 •Hf t 0 •rl 2 3 0 £ 2 O

70 £ • ft 3 •H VO 0 f t f t 0Pl 0 O 0 f t •rl ft b0 • o f t o tQ0 /—> Pi 2 £ 0 0 o 3 f t 0 o

LA ft 3 f t O f t £ 0 e* 2 2 •Hf t VO •H tQ LA 3 0 f t f t < 3 f t f tCO ON f t 0 0 tQ £ CQ tQ 0 i 0,3 f t 3 TO f t 3 o 0 < LA f t •rl f t £ vo OtQ v_»" 0 0 ft •H >_' 3 O < • 3 3 1 00 ft tQ o f t o O 2 f t o LA 0 o£ ft 0 •H 2 3 tQ •H ft f t CM o CA tQ 3CO 0 3 3 LA 0 f t £ £ + ft 2 o LA o 0

bC 3 0 O •H tQ LA f t • vo o O f t<H 0 0 b0 • f t 0 bO 3 CM 0 0 O CM f t f t 3 20 3 f t o o 0 f t £ 0 > O 0 f t o

2 3 £ f t f t r-i f t + • o f t t f t b0 •rl70 o tQ 0 70 3 3 2 £ o 0 70 30 0 ,3 3 0 f t • 3 70 O 0 CM 0 • • f t2 f t bJQ 0 VO 3 0 K •rl 0 3 bOO 70 0f t O 3 • •rl 0 0 0 ft o ft + 3 • 0 3 ft • •0 O 0 ft LA •rl 3 ft ft o £ •h r- f t 1

2 O > O f t 3 f t 0 0 LA f t 3 0 70 VO <x •rl 0 2 • £ 0 f t 3 • O •H 2 0 3 1 Oo\ f t 2 f t -ft o 0 f t tQ LA CM 2 0 f t 2 0 co f t

method, linearity of the standard plot extended to approx. 1 /«g/ml, and the extinction coefficient was calculated to be 9•9-x 10^ l.mol ^.cm (Figure 4.2). However, a fine white precipate formed within three hours, causing turbidity and falsely high extinction values.Using Slater and Sawyer’s method, the standard plot was linear up to 2.2 jxg TEP/ml (Figure 4.3), which was in good agreement with the findings of theseauthors. The extinction coefficient was (1.10 + 0.05)

5 - 1 - 1x 10 l.mol .cm and the colour was stable forseveral days.

Preliminary experiments using procedure I (Figure 4.1) failed to show an appreciable rise in malonaldehyde production in the presence of carbon tetrachloride.

-4<*-Tocopheryl acetate (2.5 x 10 M) also failed to delay the onset of spontaneous endogenous peroxidation.All subsequent experiments were carried out by procedure III (Figure 4.1'). The incubation, conditions were as follows:

5 mg of microsomal protein 8 3 .5mM KC137.2mM tris-HCl buffer pH 8.0 5 .5mM glucose-6-phosphate 0.245mM NADP+ lOmM acetamide0 . 6 units glucose-6 -phosphate dehydrogenase

Final volume 2.5 ml.

c\]

0u3b O•rlft

LAVftONvHH0bOc8PiXo0ft0c8

H(8£toOftftftO0oft+>0a0ft

-P

ft00Plaft0uft-poHft0Plc80Pl(8

-P

£VftOJo<oHLA H O Pl E

*H £ ft VD ft • ft O

Pl•rl

<ftft£CM■H

H rl rla a a

Eftu0ftft2fti-10 E1to•HPlft

Pl•rlaorHPlOftooootH■P(800-p

api

LACALA

•Pc80

0•P(80•rlHftPf"0Pl•rltoPl0•rl-P(8Pl•Hau0

•ppi0•rlO•Hftft000

H0a

pi0

•H O -P rH OPl K•rl ONLA LA O

ON

LA

to•P•HPl CO

10 ‘ft<

jxg te

trae

thox

ypro

pane

/ml

Standard

plot

prepared by

the

method

of Slater

and

Sawyer

(1971a).

0pPi id•H o8 •H

H rHo f t P 1rH •H Pi 8

< < 0 oO CQ H -p •H •EH f t O O •H

f t Pi •H 1^ • •H f t Hin u* o f t 0

id 0 8 to 0 8Pi o Pi Pi 0 •

•rl cS. o 0 a Hn- in •rl

f t VD cn f t Pi inCQ • -P in id 0 of t O id Pi •rl tH

-p •H PH H T5 cd 8 O Ks s 0 U Pi

-P T5 0 •H oo o id id -P P ■H

t • 0 0 0 K tCM CM ffi f t Q CQ rH

+

top•HCMi nvo

jxg

tetr

aeth

oxyp

ropa

ne/m

l

Trichlorofluoromethane and carbon tetrachloride were added by microlitre syringe directly to the incubation medium in 50 ml stoppered conical flasks. Incubations were for 45 minutes at 3 7 °C, unless otherwise stated. Incubations were stopped by the addition of 5 nil 10% trichloroacetic acid (TCA), giving a final concentration of 6.67% TCA, and standing the flasks in ice. The precipitate was spun down in a Mistral 6L centrifuge (MSE Ltd., Crawley, Sussex, England) at 2,000 rev./min for 10 minutes, and 2 ml aliquots of the supernatant were mixed with 2 ml portions of 0 .6 7% thiobarbituric acid (TBA), heated at 100°C for 10 minutes, and when cool the absorbance was measured at 535 nm. Incubations were carried out in triplicate or quadruplicate, and each incubation was assayed in duplicate.

4.3. Results.

In vitro incubations.The purity of the trichlorofluoromethane used was

checked by g.l.c. using the conditions described in Figure 2.8. Prior to elution of trichlorofluoromethane from the column, there were only minor disturbances in the baseline trace that at the retention time of dichlorofluoromethane accounted for something less than 0.1%.

Addition of a 1 to 500 mixture of dichlorofluoro­methane and trichlorofluoromethane to the 3 ml incubation

medium, followed by extraction of the medium with 5 ml n-heptane, resulted in a qualitative recovery of a detectable amount of dichlorofluoromethane.

However, no dichlorofluoromethane was detectable by the methods described in incubations of trichloro- fluoromethane with microsomal preparations from the following control and phenobarbital-induced animals:

male Wistar albino ratmale Wistar albino,rat (0.1% sodium phenobarbital in the drinking water for 7 days) . male C^H/Mg mouse (0.1% sodium phenobarbital) female C^H/Mg mouse (0.1% sodium phenobarbital) male golden hamster (0 .1% sodium phenobarbital) male guinea-pig (0 .1% sodium phenobarbital) female chicken.

Lipid peroxidation.Use of ethyl oleate as a solvent for the addition

of trichlorofluoromethane and carbon tetrachloride to incubations resulted in falsely high absorbance values at 5 3 5 nm» A range of solvents were tested in incubations, and ethanol, propanol, butanol and di- m e t h y l f ormamide did not form coloured products in the thiobarbituric acid reaction. Butyl acetate and liquid paraffin showed some formation of a coloured complex, but not to the same extent as ethyl oleate. Ethanol was used as solvent in subsequent experiments.

Incubation of microsomal suspensions under nitrogen prevented peroxidation for at least 60 minutes, as did storage of the suspensions at 0°C, which is shown below:

nmol malonaldehyde producedper ml incubation in G O min at

O o 037 C 37 C 0 C(under N^)

No addition 2.55 0 0.23+ 100 nl CCl^ 3 . 1 0 0

A requirement for NADPH was demonstrated by the omission of the NADPH generating system from incubations. Without NADPH, endogenous peroxidation in a 15 minute incubation was decreased from 4.37 nmol/ml to 1 . 1 7

nmol/ml suspension, and the stimulation due to the addition of 2 CCl^/ml was decreased from 26% to

1 3 % .

The extent of peroxidation during incubation atO

37 C increased with the ageing of the microsomal suspension. Using a microsomal preparation from a 2 0 0 g female rat, incubations were commenced 3 0 , 60

and 90 minutes after resuspension of the microsomal pellet. With no additions, 2.50, 2.77 and 4.l8 nmol malonaldehyde were produced per ml suspension.However, the percentage stimulation due to the addition of 1 jjJ L. CCl^/ml was 3 0 % in each case.

Tables 4.1 and 4.2 show the effects of trichloro- fluoromethane and carbon tetrachloride on lipid per-

HPB0to0

Po•HB

P•rlP0 ft•H 0-P •HP toP ft•H 0K ftO toP p0 to toft 0

H PP S H•H ' t \ Pft ■ -P T5 £•H p 0rH ft O rH

P oft 0 P ft0 H 0 -PP ft P0 £ ft 0P o•rl b[ 0

ft P <hrH O O In 0• H O 43-=F 43 0] 0 ^O Po p 0 H 0H ft £ Cd ft4> +> -p P cdp a) oft +> B H tQ

0 cd 0ft ft B too <H 042 H 43ft to 0 +>P ft i £O 0 p 0•rl pp to to cdft ft cd ftP 0

p P p0 to 0 -Hft ft toP to to to43 0 0-P p p0 ft pe X fco0 0 *Hft <H0 toft •PH rH<H P0 toft 00 PH s-/4!0•Hft•P<H0-PO0<H<HW

+ * + +o /•--N /-N *—\rH noON VPON

• o VO oo CM rH CM rH

O rH •H rH rHCn + 1 N_' w N_'rH

in rH CO cnCO r^ in O n cn

• • • • «in in m in

•to0

-ppp

•Hs

p * +•H r - -•—s ✓—s /--V X~N

o NOon “NPON

P • rH CM CM -=h0 o CM O CM ' o

•H in rH rH rH rH. m CO + 1 s » -—' v—/ '—'

P0 o ON CO CMf t ON O n ON ON rHto • • • • •P ■ <F in in into

f t0

bOP

•H0bO<

* + + +VD /•“s /--N /“N /—sO ^ - NOON

• ON CO rH r -o cn o rH rH

o rH rH rH rH+ 1 >— ' w v- 'rH cn ON CM rH

rH VO CO rH• • • • •

V0 •tf* in

H0P /—s /—s

•P /—N H rHP \ rH B B0 rH B V \0 '_'

P

B\

\t t .o o

0 CM CO•H rH N_' N_'•P v_/•H ft ft ftP cn cn cnP H H .H rHcd O O o O

O o o o0£ + + + +

•HoIn P O

H 0•P •H VP -PP O pO P p

•H •0 pf t 0•rl p CMP f t ObO O

•H 0 •to T5 o

Ptn43 V

P 0 f t43 ■P

rHto P PP P 00 O •H

•H rH toto P PP g 00 ftf t ft toto O pp toto to

H pT5 0 00 > bObO 0 PP rH PP H •H t

•rl 0 s /*ss P r

-P O rHo P cF 0)cn O >rH 0 0 •rl43 -P

'P p -P OP 0 0P 43 P f tbO P toin •rl 43 0

CO 43 -P p

iHOprHO rH ino O o• • •o o oV V Vft ft ft

* + ++ poo

s/—s ✓—\ /—s /~N

VD ONVO VD VD CM f t o cnf t f t ON ON c o CO CO

w V—* '— ' N_^ N_' N_^ W

»—s p CM VD O o O CM« m 0 O cn cn O CM O

Pi 0 -p « 1 • » • • • 1O pi CO o O o o O O

-P f t sW cd ‘cd + 1 + 1 + 1 + 1 + +E > ,3cd cn o CM cn CO VD cn

£ H 0 on ON CO CO r - c n cn -c fO rH • • t • * • • t

T5 Pi cd CM CM CM CM CM CMPl -P Ecd P!

00 otoPi f t0 0ECD

m 0 /-s /—s-P cd to ✓—\ /•—\ /—\ -'~N

pi VD VD o VPON VPOSE •d 0 rH CM c n CO ON CO f t r -0 p ! S CM CM CM CO VD in t" - c np cd w '_' '—' v / '—' v—/ N—'

<H 0Pl rH ON cn co h - CO r^- O cn r^-

CO 0 cd ON rH CM CM CO VD in n - c n£ •H £ • • • • • • • • • •0 to 0 O CM CM CM o O o o o O•H Pl ft-P 0cd ftU COcd pift co0P Hf t E

\ft 0 /—s r—\ /—\ /-scd *d 0 VP /—\ /*s /—s 'NPON VPOSE >> CO in rH CO f t •££. CM COo £ 0 Cn cn o in CO O c nna 0 0 CM CM CM ft ON CO t^ - ft vH0 TJ B v '—' >—' '_' N—' N—' v-/ '—' v-/U ■rH0 cd 0 ■ in cn ON VD VD CM ft n- rH•H Pl rH ON CM tH cn ON ON CO n- ON c ns o « cd • • • • • • • • t •

sH E o CM CM CM o O o o o rHcd

•rl EPl rH0 0

C5♦rl E-P Plcd

T i CO•H cdK0 T3 rHu 0 00 CO Pi t—i /V /—s

ft to -p ft ^~N /--N H ft ft rH0 Pl s /--N f t f t E E E E

T5 Pi 0 \ ft ft E E \ \ V V•Hft•H

ft K 0

'_'

o '_'

ti

1E E

\ V

1 1 .

\ \t .

oft

of t

o o0 • CM -=H ft CM CO•H o ■H CM »_' '_' N—'-P v--t•H f tT3 cnTJ ft ftcd oo oo015 + +

oxidation in vitro, in microsomal preparations from the rat and from the mouse and hamster. Carbon tetra­chloride produced highly significant increases in all cases, the percentage increase in mouse microsomal preparations being particularly striking. It will be noticed that the control level of in vitro peroxidation (no addition) was considerably lower in the mouse than in the rat or hamster.

The effect of trichlorofluoromethane was rather erratic. In the rat, small but significant increases were found in the experiment shown in Table 4.1 . In preparations aged for only 40 minutes, an addition of 2 j x l trichlorofluoromethane/ml suspension caused a significant 8% rise (P<0.05). 40 and 80 yfL/mladditions caused slightly greater rises. However, in the mouse and hamster microsomal preparations, addition of up to 40 j x . 1 trichlorof luorome thane/ml either had no effect or depressed the level of the endogenous per­oxidation. The only exception was found on the addition of 80 j j J L trichlorof luoromethane/ml to micro­somal preparations from the male mouse. However, the increase in malonaldehyde produced (3 8%) must be compared to the 1 3 5% increase produced by the addition of carbon tetrachloride (0.4 .l/ml).

Table 4.3 shows the effect of trichlorofluoromethane and carbon tetrachloride on microsomal preparations from male and female rats in three weight ranges, both untreated and phenobarbital-induced (0 .1% sodium

Lipid

peroxidation

in microsomal

preparations

from

male

and

female ra

ts

• /■—sto toCD 0to piPl rHcd CdPh >

+> H0

to u•H +>0 Pl£ 0

0+>Pl ft0 05h0ftft to•H cd

TJTJ

0 00 toU to

rG 0+> Plftft X0 0

'w '

Tf0 00 H vo n - CM oP cd CM o ON in

T3 B rH rHPl 0

•H f t

0Hto cd cn ON O ONS CM o O mo 0 rH rH ■H

rH ft(M1 •oo 0CO O 0 •4* IN- m NT-l pl H *4< CO CO mrcd cd rH

pi a•H

0rH in O CO COcC o ON voa rH rH

Td0 0O H CM CO

• pi cd cn 1 -cn 1T i a rH •HPl CD•ri ft0

rHto cd CO ina CM 1 cn 1o 0 rH rHtr \ ftrH1 T5o 0cn O 0 in COrH £ H O n 1 CO 1

Td cd CMpi a

•H

0H ON . incd cn 1 CO 1a rH

Td0 0O H -4 rHpi cd rH ON i oTd a rH rHPl CD

•H ft0

Hto cd ON ON Oa ON ON i rHo 0 rHt» ft1o TJvo 0O 0 o r-Pl H rH ON i vo

TJ Cd rHpi a

•H

0iH cn o ocd cn o i ONa rH rH

<--NrH

r sH

a a\ ft \ ft

-4 i cnrH tC

HO rH O OO o s«» O

cn

B\%rH

ftcnHOO

Ha\o

phenobarbital in the drinking water for 7 days).The results are expressed as a percentage of control values because endogenous peroxidation decreased as the age (and weight) of the rat increased. It can be seen that In most cases addition of trichlorofluoro­methane to the incubation medium depressed lipid peroxidation, whereas carbon tetrachloride caused increased peroxidation. Phenobarbital pretreatment of the animals caused increases in endogenous peroxidation of between 10% and 40%. However, only phenobarbital- induced rats of 1 3 0 - 1 5 0 g showed larger increases in peroxidation due to carbon tetrachloride compared to controls.

4.4. Summary.No dichlorofluoromethane was detected as a

metabolite of trichlorofluoromethane in incubations of microsomal preparations from the rat, mouse, hamster, guinea-pig or chicken. The limit of detection of dichlorofluoromethane in the presence of a large concentration of trichlorofluoromethane was approximately 0.2%.

The effect of trichlorofluoromethane on microsomal lipid peroxidation in vitro was variable, in some experiments (e.g. Table 4.l), lipid peroxidation was significantly increased by addition of trichloro- fluoromethane (1 - 80 to the incubation. In otherexperiments (Tables 4.2 and 4.3) peroxidation was

unaffected or even depressed by the addition of trichlorofluoromethane. Carbon tetrachloride was used as a positive control; in all experiments, carbon tetrachloride produced significant increases in the in vitro lipid peroxidation when added to the incubation medium at a concentration of 1 yjCL/ml. In the mouse, carbon tetrachloride produced greater percentage increases in microsomal lipid peroxidation than in other species, but trichlorofluoromethane caused decreased levels of peroxidation.

CHAPTER 5

PRELIMINARY TOXICITY EXPERIMENTS.

Contents.

Introduction.

Methods.Preliminary toxicity experiments tissue preparation and protein estimation; lipid extraction; quantitation of total lipid.

Results.Estimation of lipid by the colorimetric method; storage of lipid extracts; preliminary toxicity experiments..

Summary.Lipid assays; preliminary toxicity experiments.

5.1. Introduction.Chlorofluoroalkanes have been reported to be of

low toxicity, judged by both acute and chronic exposures.To confirm these reports, preliminary toxicity experiments were designed-to investigate the effects of high dose levels of trichlorofluoromethane and 1 ,2 -dichloro-1 ,1 ,2 ,2- tetrafluoroethane on body and tissue weights over a period of days or weeks. In addition tissue lipids were measured, as 1 fatty liver1, caused by the accumulation of triglycerides, is a clear toxic effect in carbon tetrachloride poisoning.

5.2. Methods.

Preliminary toxicity experiments.Four experiments were carried out:

(i) 2 male and 2 female Wistar albino rats (Porton strain) were dosed with trichlorofluoromethane (400 mg/kg body weight) i.p. daily for 35 days.Control rats (2 male and 2 female) were dosed an equivalent amount of ethyl oleate. Body weight was measured daily and liver, kidney and lung weights at the end of the experiment. Protein and lipid were assayed in these tissues.

(ii) 2 male Wistar albino rats were dosed with trichlorofluoromethane (1 . 9 g/kg and 2 . 5 g/kg body weight) i.p. for 5 days. 2 control rats received

ethyl oleate. Body and organ weights were measured on the sixth day, as were tissue lipids. Serum, liver and adipose tissue were assayed qualitatively for

, trichlorofluoromethane by g.l.c.

(iii) 4 groups (randomly selected) of 6 male Wistar albino rats were treated as follows for 37 days:

(a) ethyl oleate alone, i.p.(b) trichlorofluoromethane in ethyl oleate

(400 mg/kg), i.p. •(c) trichlorofluoromethane in ethyl oleate +

sodium phenobarbital in 0 .9% saline ( 3 0 mg/kg)

i.p.(d) sodium phenobarbital in 0 .9% saline ( 3 0 mg/kg)

i.p.

Body and tissue weights and liver protein and lipid were measured.

(iv) 2 male Wistar albino rats were dosed1 ,2 -dichloro-l,1 ,2 ,2-tetrafluoroethane ( 1 5 0 mg/kg) i.p. daily for 17 days. Body and tissue weights were measured.

Tissue preparation and protein estimation.Tissues were homogenised in 3 volumes of ice-cold

0.25M sucrose, as described before (4.2) and the homogenate was fractionated following the general principals of de Duve et al. (1955).

Protein was measured by the method of Lowry et al. (l9 5l). 5 nil freshly prepared copper reagent (2%Na^CO^, 0 .02.% sodium potassium tartrate, 0 .0 1% CuSO^.^H^O was added to 1 ml of a suitable dilution (in 0.5M NaOH) of each fraction and mixed well. After 10 minutes,0.5 nil of Folin-Ciocalteu reagent (l:2 (v/v) Folin reagent : distilled water) was added and immediately mixed. The absorbance was measured after 30 minutes at 720 nm. Standards were prepared with crystalline bovine serum albumin and the standard plot was linear up to 200 yug/ml.

Lipid extraction.Lipid was extracted by the method of Folch et al.

(1937). Tissue was homogenised in 2:1 chloroform- methanol mixture and the final volume adjusted to 20

times the volume of the tissue used. This was filtered through glass-fibre filter discs and mixed well with 0.2 times its volume of 0.29% NaCl. This was centri­fuged at 2,000 rev./min for 10 minutes (MSE Mistral 6L) to separate the phases. The upper phase was carefully sucked off and the interface was washed three times with a chloroform-methanol-water mixture (3 i ^ S : k 7 by volume). The lower phase and remaining traces of upper phase were made into one by the addition of a small volume of methanol. Chloroform-methanol mixture(2 : 1 v/v) was added to the washed extract to make it up to any desired volume.

Quantitation of total lipid.Total lipid was measured gravimetrically.Initially, 1 ml and 2 ml aliquots of the washed

chloroform-methanol extracts were evaporated down in weighed test tubes and the residue weight obtained by difference, after drying the residue for 2 hours at 105°C. «

The accuracy of pipetting 1 ml aliquots was approximately + 0.4%, as estimated by the weights of six aliquots (1.2457 + 0.0055 g)» The accuracy of weighing empty tubes was also good (l8 . 8 9 5 3 + 0 . 0 0 0 1 g). However, from results obtained for residue weights, it was apparent that the accuracy of weighing a difference of approximately 2 mg was insufficient to obtain reasonable triplicates.

The final modification used was a major simplif­ication of the method of Sperry and Brand (1955)• A 15 ml glass-stoppered tube was heated at 90°C for 2 hours, and weighed when cool. Two 5 ml aliquots of the chloroform-methanol extract were evaporated down in this tube at 55“60°C under a stream of oxygen-free nitrogen. The tubes were desiccated for 3 days over cone. H^SO^, reweighed, heated at 9 0°C for 2 hours and finally reweighed when cool. The weight of the residue

Oafter drying at 90 C was decreased by an average of 0.15 + 0.06 mg. Control tubes run through the full procedure showed an overall increase in weight of 1.13 + 0.09 mg. Test values were of the order of

50-75 mg.A second method was used to quantitate lipids.

Frings and Dunn (1970) reported a colorimetric method for the determination of total lipids in 0 . 1 ml serum, which was based on the formation of a sulpho-phospho- vanillin complex involving a carbon-carbon double bond in some way. These workers used olive oil dissolved in ethanol as a standard, and stated that the colour was "stable for at least 10 minutes". They also stated that "triolein, oleic acid, linoleic acid, linolenic acid and cholesterol reacted quantitatively in the method", but did not explain this in any way.The method was open to criticism on several points. Firstly, the assay required the transfer of an aliquot (0 . 1 ml), containing 9 5 % cone. H^SO^. 0 . 1 ml conc. HgSO^can only be pipetted with great difficulty, yet no details were given of the procedure. Secondly, olive oil is not homogeneous and the composition of different batches may alter, thus affecting the standard curve. Thirdly, without further information on the ' nature of the chromophores formed with different compounds, it seems unjustified to equate mg olive oil equivalents with mg lipid.

The method was adapted for use on chloroform- methanol tissue extracts. Oleic acid was used as astandard, dissolved in chloroform-methanol (2 : 1 v/v) and stored at 4°C. Phosphovanillin reagent was prepared by adding conc. phosphoric acid ( 8 0 0 ml) with

constant stirring to 0 .6% (w/v) vanillin solution( 2 0 0 ml) and was stored in the dark at room temperature.

2 ml conc. H^SO^ was pipetted into a tube containing 0 . 1 ml of a chloroform-methanol extract or standard. This was mixed on a Whirlimixer (Fisons Scientific Apparatus Ltd., Loughborough, Leics., England). The resulting emulsion was heated for exactly 10 minutes on a boiling water bath, and then cooled for 5 minutes in ice-cold water. A 0.1 ml aliquot was transferred to a second tube, using a fixed-volume microlitre pipetter, with a disposable tip. 0 . 1 ml chloroform-methanol mixture was used as a blank. Phosphovanillin reagent (2.5 ml) was added and well mixed. The samples were incubated at 37°C for exactly 15 minutes and were allowed to cool for 10 minutes at room temperature. The absorbance was measured at 540 nm in 1 cm path length cells using a Unicam SP500 spectrophotometer. Triplicate analyses were made of each sample. Results have been expressed as mg oleic acid equivalents.

The stoichiometry of the colour reaction was investigated by comparison of oleic, linoleic and linolenic acids and cholesterol as standards.

5.3. Results.

Estimation of lipid by the colorimetric method.Figure 5«1 shows a comparison of oleic acid

Figure 5 .1.

Standard plot for colorimetric lipid assay.

Absorbance (units)0.6

0.5

0.2

Dissolved inethanol

a chloroform-methanol (2 : 1 v/v)

1 2 3 4 / 5 6mg oleic acid/ml

standards prepared in ethanol and in chloroform - methanol (2:1 v/v). It is apparent that the standards prepared in ethanol show a greater degree of variation within triplicates than those prepared in chloroform- methanol and they also show poorer linearity.

Figure 5»2 shows standard plots prepared with oleic, linoleic and linolenic acids and with cholesterol.These compounds contain 1, 2, 3 and 1 double bondsrespectively. It can be seen that the standard plots lay close together and that cholesterol and linolenic acid shared exactly the same plot. If, however, the extinction coefficients are compared, cholesterol and linoleic acid then show the same value. This is shown below:

No. of Absorbance at Extinctiondouble bonds lmg/ml (units) coefficient

(l.mol.*cm ^)

oleic acid 1 . 0 . 8 1 0 2 . 2 8 X 1 0 5

linoleic acid 2 0.875 2 . 5 6 X 10 5linolenic acid 3 0 . 6 5 0 1 . 6 5 X 1 0 5

cholesterol 1 0 . 6 5 0 2 . 3 6 X 1 0 5

Storage of lipid extracts.Two portions of a crude extract of liver lipid

/ O \were stored, one at room temperature (22 C) and one at4 C. On day 1 , day 4 and day 8 , aliquots were taken and washed as described above (5*2). The washed extracts

Figure 5 - 2.Comparison of unsaturated Fatty acids as standards in

the colorimetric lipid assay.Absorbance (units)

3

0.2

oleic acid1linoleic acidlinolenic acidcholesterol

kmg/ml

were assayed by the colorimetric method. The washed extracts prepared on days 1 and 4 were retained and assayed again on days 4 and 8 and day 8 respectively. The results are expressed as mg oleic acid equivalents/ g liver + S.E-.M. No significant decrease was found.

sample storedat

Day 1 Day 4 Day 8

A1 4°C 2 4 . 7 2 + 0 . 5 7 2 3 . 6 0 + 0 . 7 0 2 6 . 4 3 + 1 . 0 0

A4 4 ° c - 2 3 . 8 3 + 0 . 6 7 2 3 . 5 3 + 1 . 1 5

A8 4 ° c - - 2 6 . 6 2 + 1 . 0 0

B1 22 °C 2 3 . 7 5 + 0 . 3 1 2 3 . 8 7 + 0 . 8 7 2 5 . 3 2 + 1 . 5 2

b 4 to to 0 O - 2 3 . 7 0 + 1 . 8 6 2 5 . 4 8 + 0 . 8 6

b 8 to to0 O - - 2 2 . 5 3 + 0 . 9 3

Preliminary toxicity experiments.

(i) Table 5.1 shows the results of this experiment. No differences in protein content of the various sub- cellular fractions were found. Slight alterationswere indicated in liver and lung total lipid, but the values were not significantly different. No change was found in kidney protein or lipid.

(ii) Table 5-2 shows the results of this experiment. There was a significant increase in liver weight in the test animals. Total lipid levels were not altered in liver, kidney, lung, serum or adipose tissue.

Traces of trichlorofluoromethane were found in

in<D

H

dft

ft•H * * *

to o COto ✓-S bO n- n-in ft d 1 i I i • •10 A 2 o on bO r—1•rlin 0m £U in0 73ft 0inH bO •to•H tn O o 00 rH 00 rH rH<0 O 0 -cf cnT3 O Pi • • • • t • • •

rH 'd O o o o o o o o\ •rl

bO bO r* i

\bO •g -pO iso 02 Uto 0 o in o o o o o m(1) to !> m . VO ' cn in CM CM rH CMPi •H •H • t • • • • • •(0 ft i—1 cn <M cn cn cn cn c n c nrdft0g0u0 . dd 0H > '“Sft 0 bO0u d O cn in ON rH ON ON0 •H to ' O n- in rH CM rH rHH (0 in rH CM rH rH rH rH rH rH£ to dO 'd•H •d ft inft is cn

ft /0 t

ftft iso0 H /—s VO n- cn CO rH VO VO chft »d bO in VO vo m VO in in inft •rl0 ft•rl0 dHftl

✓—s•rl 0 0 0 0N_' ft ft ft ft

ft Pi (0 d d dft d •H 0 0 d 0 0£ 0 l—l r—1 •H i—l H0 g ft 0 O 0 0g ■p cn ft•rl (0 i—l 1—1 r—1 cn i—i i—1u 0 o in in H0 Sh o rd rd O rd £ft ft ft ft o +> ftK 0 0 0 00inft•rl 0 0O H ' H■rl 0 0 (0 dK M H H g g0 0 d d 0 0ft S S ft ft

lungs

pooled

from

two

animals.

/—N-p£ w CV] CO O ONbO bO vo A- a a-•H pi • • • •CD J3 o o 00

£ HkA730rQbOO toO >n CA in ON rH■H 0 CO CA A- CO\ Pi • • • •bO 73 o o o o'— *H•co-p£o2 JhCO 0 IN. VO rH CMco > ON ON CM CA•H -H • • ' • •EH H -tf* IA •it* -it*

U0>O bO

•H CO CO LA -4* COcd >a CA CA CAbO CCS73•

-P VO£•

•P£H CA CM VO O(d bO a a A- A-•H w rH rH rH rH■P•HPIH

<H0U0 0 bOCQ > w A rH0 0 VO rH

73 CO • • 1 1fa £a rH CM

H c a (dcd H 73+> O0 O IAEh

0 0•P -P-P cd cdP! Pi <D 00 •H H H -g 0 0•p fa<d CA H rH0 H !>>u O rG £Eh o +> ■P0 fa

the serum and liver and greater concentrations were found in adipose tissue.

(iii) The results of body and tissue weights in this experiment are shown in Table The onlysignificant difference found (in liver weight expressed per 100 g body weight) was not attributable to tri- chlorofluoromethane, but to the phenobarbital treatment. The liver lipid values are given in Table 5-^- Total lipid measured gravimetrically and lipid measured colorimetrically showed no significant differences between the three groups receiving ethyl oleate.However, the group of animals receiving on3.y pheno­barbital (d) showed significantly lower lipid levels measured by both methods compared to those in the ethyl oleate control group (a).

Subtraction of mg oleic acid equivalents from mg total lipid showed a significantly lower difference (P < 0.05) in animals receiving trichlorofluoromethane compared to animals receiving solvent alone. It should be noted that the difference between total lipid and oleic acid equivalents for group (d), which did not receive ethyl oleate, was the same as for the control group (a).

(iv) In the rats dosed with 1 ,2 -dichloro-1 ,1 ,2 ,2- tetrafluoroethane ( 1 5 0 mg/kg) for 17 days, body weight gain was slightly depressed. However, one test animal contracted peritonitis toward the end of the

cnin0)rHAd

-pAbO•HCD£OPmm•H

-p

'dPicdInK0APiobO

bO£oO

oP!cd£-pd£oPopirH<HOPiOrHAO•HPi-P

OCQ-POd«H<h

-pPid£•HPdftKd

•p•HO• HK0

o■

o■»4 1o

V.VIo• • • •

P o o o O•rHcd + 1 + 1 +1 + 1P,Q VO VO N N-

VO IN VO VO• • • •

o o o o

/—\-pA CO o COb0 o rH o o•H • • • •d W o o o o£ bO

P + 1 + 1 + 1 + 1P

U H in CM ON CO0 VO in VO •

A • • • •o O o . o

bO

OOrH\ CO CO CObO o o o . o

V-/ to • • • •tn o o o o

• d+> P + 1 +1 + 1 + 1£ T3

•rl ON rH o Ond VO VO N- VOP • • . • •m o o o Om

•H

VO oo rH rH rH

• • • •P o O o od> + 1 +1 + 1 + 1•HH On CO o VO

CO ON n- rH• • • •

cn cn

P•H bO

m -CM VO VOp . CM CM rH rH

•rl mcd + 1 + 1 + 1 + 1bO cd

73 n- cn CM o• cn in

-P n- rH rH rH rH£ cn

rH r—\ CM Ocd bO rH rH CO ON

•rl-P + 1 H* 1 + 1 + 1•H •P ■P cn O CM cnH £ rH rH ON orH rH rH

CO<H H0 cd

£ VO m• •rH

0 P£ cd

PiP0 cd A 0 13PCD

d-PcddrHOH

, P

d-PcddHOrH

-P-Pd

P♦ HdPcd, P-pd£oPiOPH<HOPiOrHAo•rHp-p 'bO

\bO£

Hcd-p•rlAPcdAOPd£PibOA\bO

Ocn

dPcdA-pd£oPiOPrH<HOPOrHrPo•rlp-pbD

\bO£

Hcd-p•rHAPcdOPd£PibOA\bO£

-P O O Od cn

cd A O u

Pi Pi Pi PiP P P P0 0 0 0P P P PCD CD CD CD

Toxicity

experiment

(iii)

- liver

lipid

levels.

(8>

•rlPi

LA rH O4+ * O t8 0

0 on ■if rH • fto • • • • O •rl TDPS rH •H rH CA & •rlCD b/) V Ph OPh + 1 + 1 + 1 + 1 ft C8 c8<D bO rQ

f t CA 03 LA rH 0 oft • • • • 0 Pi •rl•H o IA CA CA u 0 0Q ca 03 CA CA ft A H

PS ft 00o bO TJ

M PSa \ c8O bOPi £ T3

/—N ft •rlbO o ft£ to CA •rl

Pi rHc—■. 0 +to p ft rHft 0 ft 0 0 c8PS £ •rl f t -P f t0 •p P (8 (8 0

H 0 CO ON ON VO 0 0 .ftc8 a • • • • * rH rH> 03 03 o rH O 0 PS

•H o 0P •rl + 1 + 1 + 1 + 1 H rH 0O’ Ph Pa IA £<D ft CO VO rH £ ft

0 • • * • -P -P 0a r - CO IN. 0 0 A

•H •H ca CA CA 03O Ph Pi Pi to(8 0 •rl •rl 0

rH oO 0 0 0 Pi

•H O PS Pi 0<D ■-/ c8 c8 Ph

H £ £ 0O -p -P ft0 0 ft£ £ •rlO 0 nPh Pi0 0 • rH

/—s P* P* rH <8rH r—1 rH c8 Pi/—s /—\ 0 f t f t f t TD

bO P Pi O 0 •rl •rl\ 0 -P Ph Ph ,Q >bO £ Pi O 0 Ph •rH£ f t 0 rH rH c8 TJ

0 CO r - .ON 03 o £ A A Pi£ _ • • * • v-^ O 0 0 •rl

rH CA rH CA •rl •rl PS•rl o 0 Ph Ph 0 0f t •rl + 1 + 1 + 1 + 1 -p -P f t £ A

•rl u (8 f t ftrH -p CA rH CO Is* 0 bO bO

0 • • • • rH M A bO f trH a CO o [N. O \ \ M 0(8 •rl VO VO IA bO bO \

f t > H a a bO •O c8 lA a S

f t Pi rCl O o •bO -P o o o m

>w/ 0 CA •w

to + ift H <8 & o T50 <8 PS

£ VO VO VO VO c8• •rH ft ft ft ft 0

0 PS 3 pi pi PJ a& (8 0 0 0 0

Pi p. Ph PhCD CD CD CD +

ftrlP0 c8 HP o PPhcd

experiment and the results (Table 5-5) must therefore be interpreted with caution.

5* :. Summary.

Lipid assays.The gravimetric procedure for assay of total lipid

that was finally adopted proved to be reproducible, and was a great simplification of the method of Sperry and Brand (1955)•

The colorimetric procedure, used by Frings and Dunn (1970) for the assay of total serum lipids, appeared to involve the reaction of the sulpho-phospho-vanillin complex with compounds containing double bonds.However, the formation of the coloured complex was not dependent on the degree of unsaturation in individual molecules. The slight differences in extinction coefficients for the unsaturated compounds tested was probably due to differences in the chromophores formed.The colorimetric assay appeared to be a measure of the molecules containing at least one C=C bond. Results could thus be interpreted either as a measure of unsaturated lipids or as a reflection of the content of unsaturated residues in a particular population of lipids.

Oleic acid was chosen as the standard in the assay for two reasons: it is a pure compound and its standard plot lay betwen those of the other compounds tested, whether expressed as mg/ml or umol/ml. Use of

-Toxicity

experiment

(iv) - The

effect

of 1,2-

dich

loro

-1,1

,2,2

-tetrafluoroethane

(150

mg/kg)

ft•HCOcdTJt-■H

OftH•HcdTJ

ONCO VOin

cnvo

co

VO

cn vo t^. in

ONCM in

cn

cn■H>NcdTJS0Phft

-prSto

•rl0£

-PtooTJ£cd

to•H-P•HO

-P•Hu0ft*00POcdPh-p

oorHcdS

•HPicd

chloroform-methanol (2 : 1 v/v) as the solvent instead of ethanol improved the reproducibility and linearity of the standard plot up to 6 mg oleic acid/ml. The linearity reported by Frings and Dunn (up to 10 mg olive oil/ml) could not be reproduced.

Storage of lipid extracts was possible, even at room temperature, for several days. No decrease was found in the stored lipid extracts when the lipid was measured colorimetrically, which suggested that autoxidation (Dodge and Phillips, 1966) did not occur.The slight increase at 8 days was probably due to small losses of solvent.Preliminary toxicity experiments.

Intraperitoneal injection of large doses of trichlorofluoromethane or 1 ,2-dichloro-1 ,1 ,2 ,2 -tetra- fluoroethane produced little effect over several weeks. Tissue proteins and total lipids were unaffected. ’Unsaturated lipid’ in the liver was not significantly affected. However, the difference between total and ’unsaturated1 liver lipid ('saturated’ lipid) was significantly lower (P<Q.05) in animals receiving trichlorofluoromethane in ethyl oleate compared to ethyl oleate controls.

The qualitative analysis of tissues in experiment(ii)!indicated that trichlorofluoromethane was still present in the tissue 24 hours after the final dose,of a series of 5 repeated doses (1 . 9 g/kg).

CHAPTER 6

INTERACTIONS WITH CYTOCHROME P-450 AND DRUG-METABOLISING ENZYMES.

Contents.

6.1 Introduction.

6.2 Methods.Animals; tissue preparation; extraction of microsomal preparations with isooctane; digestion of microsomal preparations with phospholipase C; digestion of microsomal preparations with phospholipase D; measurement of cytochromes; binding spectra; conversion of cytochrome P-450 to P-420 during incubation with trichloro- fluoromethane; hexobarbital sleeping times; zoxazolamine paralysis times; in vitro hepatic enzyme assays.

6.3 Results.Spectral interactions;determination of K values;seffect of isooctane extraction; effect of phospholipase digestion; interference by chlorofluoro- alkanes with type II spectra; reduced spectra; effect on measurement of cytochromes b^ and P-450; destruction of cytochrome P-450 by incubation at 3 7 °C; effects on hexobarbital sleeping times and zoxazolamine paralysis times; in vitro enzyme assays.

6.4 Summary.

Pagel6 l

163

180

217

6.1. Introduction.The effect of chlorofluoroalkanes on the metabolism

of foreign compounds is of importance as considerable use is made of them in aerosol sprays containing drugs, pesticides and food additives. These compounds, most of which are foreign to the mammalian system, are metabolised principally by the hepatic microsomal mixed-function oxygenase system, the terminal electron acceptor being cytochrome P-450.

In addition, as already suggested . (.1.12), the hepatic microsomal electron transport chain would be a probable site for metabolism of the chlorofluoroalkanes, if this occurs. Interaction of chlorofluoroalkanes with this system, and particularly with cytochrome P-450, might indicate the possibility of metabolism.

Two aspects of the interaction of the chlorofluoro­alkanes with drug metabolism were considered:

(a) The interactions of several chlorofluoro­alkanes and chlorinated hydrocarbons with hepatic cytochrome P-450 were investigated spectrophotometrically to determine the class of spectrum produced with oxidised preparations, the apparent spectralconstant (K^) and the nature of any binding in chemically reduced (Na^S^O^) microsomal suspensions. The spectral interactions of several of these compounds and some chlorinated hydrocarbons were studied in several species.

(b) The influence of trichlorofluoromethane, in

particular, on in vivo and in vitro hepatic drug metabolism was studied, and compared with the effects of carbon tetrachloride. In vivo studies were of hexo­barbital sleeping times and zoxazolamine paralysis times. To determine the effects of chlorofluoroalkanes in vitro on drug metabolising activity, four compounds showing type I spectral interactions and four showing type II interactions were used as substrates. The reactions tested were:

Type I

(i) biphenyl 2- and 4-hydroxylation(ii) ethylmorphine N-demethylation(iii) hexobarbital oxidation(iv) p-nitroanisole O-demethylation

Type II

(v) acetanilide 4-hydroxylation(vi) aniline 4-hydroxylation(vii) thiabendazole hydroxylation(viii) zoxazolamine hydroxylation.

The effect of carbon tetrachloride on these activities was used as a positive control.

6.2. Methods.

Animals.The animals used in these experiments were as

described before (4.2) and were maintained under the same conditions.

When required, animals were pretreated as follows:

(i) Sodium phenobarbital (0.1%) in the drinking water supply for 7-l4 days, or by i.p. injection.( 1 0 0 mg/kg body weight) daily for 3 days.

(ii) 3-Methylcholanthrene ( 3 0 mg/kg body weight) in ground nut oil daily for 3 days by i.p. injection.

(iii) Carbon tetrachloride (0.2 ml/kg body weight) in ethyl oleate daily for 7 days by stomach tube.

(iv) Trichlorofluoromethane (l g/kg body weight) in ethyl oleate daily for 7 days by stomach tube.

Controls were given 0.9% saline i.p. or ethyl oleate by stomach tube as appropriate.

Tissue preparation.Whole liver homogenates were prepared as described

previously (4.2).For enzyme assays, the homogenate was centrifuged

as before (4.2) and either the 11 0 , 0 0 0 g supernatant1 or

the microsomal pellet, resuspended in buffer appropriateto the assay, was used.

For binding studies, the liver homogenate wascentrifuged at 12,000 g for 20 minutes at 4°Ca v •(High Speed l8 centrifuge, 8 x 50 ml rotor, MSE Ltd.)to ensure that the supernatant was free from contaminationby the mitochondrial or lysosomal fractions. Thesupernatant was very carefully decanted and wasrecentrifuged at 4°C in a Superspeed 50 centrifuge(MSE Ltd.) either at 104,000 g for 1 hour (8 x .25 mlav •rotor) or at 1 5 7 , 0 0 0 g for 40 minutes ( 1 0 x 10 mlav •rotor). The supernatant was discarded and the microsomal pellet was resuspended with a motor-driven Potter homogeniser in fresh 1.15% KC1 and recentrifuged under the same conditions. The supernatant was discarded again. This procedure reduced contamination of the microsomal pellet due to haemoglobin and reduced the concentration of soluble enzymes and cofactors trapped in the pellet. The 'washed1 microsomal pellet was either resuspended in approx. 1 ml O.IM Na^HPO^.NaH^PO^ (O.IM phosphate buffer) pH 7 •6 per g wet weight of original tissue and used immediately or' was stored for up to 24 hours at 4°C, overlaid with the phosphate buffer.

The microsomal suspension (approx. 20 mg protein/ ml) was, diluted with buffer as required, to between 1 . 0 and 4.0 mg protein/ml, measured by the method of Lowry et al. (1951) (see 5*2).

Recovery of microsomal protein in the 'washed' pellet was 64.0 + 0.5% (3 experiments).

Extraction of microsomal preparations with isooctane.The 'washed* microsomal pellet was extracted by

a modification of the method of Leibman and Estabrook (l97l). The pellet was resuspended in 0.25M sucrose, containing 50mM Tris-HCl, pH 7*^, to give a protein concentration of approx. 30 mg/ml. The suspension was stirred at 4°C with 0 . 6 7 volumes isooctane (2,2,4-tri- methylpentane, spectroscopic grade) for 25 minutes.The resulting emulsion was centrifuged at 104,000 g for 1 hour. The isooctane and aqueous layers were removed and the microsomal pellet resuspended in 0.1M phosphate buffer, pH 7*6.

The recovery of microsomal protein and cytochrome P-450 reported by Leibman and Estabrook (l97l) was 19%, using different centrifugation conditions. The modified

procedure described above gave 7 6% and 103% recovery of protein in two experiments.

Digestion of microsomal preparations with phospholipase C .Microsomal preparations were treated with phospho­

lipase C (from Cl. welchii; supplied by Sigma Chemical Co. Ltd., London, England) by the methods of Chaplin and Mannering (1970) and Eling and DiAugustine (1971K Microsomes (5 mg protein/ml) were incubated with stirring at 22°C with phospholipase C (l.O mg/ml) in 0.02M Tris-

HC1 buffer pH 6.5, lmM CaCl for 40 minutes. The reaction was stopped by adding EGTA (ethyleneglycol- bis-(j3-aininoethylether )N, N*-t etraacetic acid) (final concn. 5mM). The digested suspension was centrifuged at 1 5 7 , 0 0 0 g for 40 minutes, and the pellet resuspended in O.IM phosphate buffer, pH 7-6.

Digestion of microsomal suspensions with phospholipase D.The method of Eling and DiAugustine (l97l) was

used, except that pH 6.5 Tris-HCl buffer was used for the incubation. Conditions were exactly as described for phospholipase C. Phospholipase D (from cabbage) was supplied by Sigma Chemical Co. Ltd.

Measurement of cytochromes.Cytochrome P-450 was measured by the method of

Omura and Sato (1964a), using a Unicam SP 1800 dual beam recording spectrophotometer with 1 cm path length cuvettes.

Microsomal suspensions (protein concentration 1-4 mg/ml) were placed in both sample and reference cells. The contents of the sample cell were bubbled through with CO for 20-30 seconds and then the contents of both cells were reduced by the addition of a few mg of solid N a S Or. After mixing, the difference spectrum2L Ca frwas recorded between 390 nm and 500nm. Cytochrome P-450 was estimated in terms of units of absorbance difference between 450 nm and 490 nm.

Cytochrome P-420 was estimated in a similar fashion, by the difference in absorbance between 420 nm and 490 nm.

Extinction coefficients for these cytochromes inrabbit microsomal preparations (Omura and Sato, 1964b)were used to calculate concentrations. For cytochrome

1 — 1P-450 and cytochrome P-420 they were 91 l.mmol .cm -1 -1and 111 l.mmol .cm respectively.

Cytochrome b^ was estimated by reducing the samplecell contents with NADH and recording the spectrumbetween 390 and 500 nm. The extinction coefficient

- 1 - 1between 424 and 409 nm was taken to be 185 l.mmol .cm (Omura and Sato, 1964a).

Binding spectra.Five ml of a microsomal suspension (1-4 mg protein/

ml) in O.IM phosphate buffer pH 7-6 was divided between two 2 . 5 nil stoppered cuvettes, and the cuvettes placed in the forward sample position of a Unicam SP l800 recording spectrophotometer. A baseline of the optical difference between the two cuvettes was recorded between 350 nm and 460 nm. Substrate (in ethanol) was added to the sample cuvette by microlitre syringe and an equal amount of pure ethanol was added to the reference cuvette. After mixing, the optical difference between sample and reference was recorded over the base­line. For measurement of absorbance changes, the spectrum was scanned at 4 nm/sec at a slit width of 3 nm.

Scan speed and slit width were reduced to 1 nm/sec and 0.5 nm respectively for the determination of the wavelengths of spectral maxima and minima.

The spectral, binding constant (Kg) was determined by measuring the absorbance change in the difference spectrum at a range of substrate concentrations. The total absorbance change produced at each substrate concentration was taken to be the sum of the absorbance differences between the peak and baseline and the baseline and trough.

Sequential additions of substrate were usually made to the same microsomal suspension in the sample cuvette. The volume in the reference cuvette was adjusted with solvent (ethanol). In a few cases, fresh microsomal suspension was used to determine the absorbance change at each substrate concentration.The maximum volume added to the cuvette was 5 0 J J ^ L i.e.2%. The range of substrate concentrations used was usually 1/3 to 3 times the Kg value.

The results were plotted in one of two ways. A plot of the reciprocal absorbance change (l/AE) on the y axis versus reciprocal substrate concentration (l/S) on the x axis yielded a normal Lineweaver-Burk plot (Lineweaver and Burk, 1934). The intercept on the x axis was -l/K . An alternative plot (Hanes, 1932)

S

was also applied to the determination of Kg. The substrate concentration divided by the absorbance change (S/AE) (y axis) was plotted against substrate

concentration (S) (x axis). K was determined directlysfrom the intercept on the x axis. Plots were fitted to the experimental points by the method of least squares. In the case of plots with a break in the straight line, the two lines were initially fitted by eye.

Spectra were also recorded between 390 nm and 500 nm after chemical reduction of the contents of both cuvettes with a few mg solid Na^S^O^. The baseline Obtained with reduced microsomal suspensions was level between 430 and 500 nm. These spectra will be referred to as reduced spectra.

In one experiment, an NADPH generating system was substituted for Na^S^O^.

Conversion of cytochrome P-450 to P-420 during incubation with trichlorofluoromethane.

The in vitro system reported by Heni (l97l) was used. Composition of the incubation medium (final volume 5 nil) was as follows:

10 mg microsomal protein in 0.05M Tris-HCl buffer,

pH 7.5.0.3mM NADP+.1.5 units glucose-6-phosphate dehydrogenase.1.5mM glucose-6-phosphate.ImM MgClg.

In some incubations the buffer contained 20% (v/v)

glycerol.The incubation tube was flushed with oxygen,

stoppered and pre-incubated at 37°C for 5 minutes. Trichlorofluoromethane or carbon tetrachloride was then added with a microlitre syringe. Incubations were stopped at different time intervals up to 30 minutes by plunging the tubes into ice. When cold, the incubation medium was poured into two cuvettes and the content of cytochrome P-450 determined as described above.

Hexobarbital sleeping times.Sodium hexobarbital (in 0.9% saline) was injected

i.p. into female rats at a dose level of 100 mg/kg body weight at specific time intervals after dosing the animals orally with chlorofluoroalkanes. The time interval chosen was designed to allow the blood level of chlorofluoroalkane to reach its maximum and was 10 minutes for trichlorofluoromethane and 30 minutes for 1,1,2-trichloro-1,2,2-trifluoroethane and dichloro- fluoromethane.

The sleeping time was taken to be the interval between loss and regain of righting reflexes.

Zoxazolamine paralysis times.Zoxazolamine (2-amino-5-chlorobenzoxazole),

purchased from K & K Laboratories Ltd., Plainview,N.Y., U.S.A., was injected i.p. into female rats at a dose level of 7 0 mg/kg body weight, as described for

hexobarbital sleeping times.Paralysis time was taken to be the time interval

between loss and regain of reflexes.

In. vitro hepatic enzyme assays.All incubations were carried out in duplicate in

stoppered tubes at 37 °C, and were shaken at 100 cycles/min.All centrifugations were carried out using a

Mistral 6L (MSE Ltd.) at 2,000 rev./min for 10 min.The following abbreviations are used: G-6-P

(glucose-6-phosphate), G-6-P dehydrogenase (glucose-6- phosphate dehydrogenase), Tween 80 (polyoxyethylene sorbitan mono-oleate).

(i) Biphenyl 2- and 4-hydroxylation.A modification of the method of Creaven et al. (19&5)

was used to determine the formation of 2- and 4-hydroxy- biphenyl .

Incubations were set up as follows:

Test Blank 4-0H 2-OHstandard standard

(volumes in ml)

Microsomes O • to CMt

o o • to o • to

Buffer 0.7 0.7 o • to o • to

Cofactor solution o t to o • to O • to 0 . 2

lOOmM MgSO^ 0 . 1 0 . 1 0 . 1*

0 . 1

4-0H standard - - 0.5 -2-OH standard — — ■ — 0.5

* *Substrate 0.5 0.5 0.5 0.5Final volume 1.7 nil. * added after incubation.

Microsomes were suspended at 10 mg protein/ml.Buffer was O.IM Na^iPO^. Nal^PO^ , pH 7.6.Substrate was 15mM biphenyl in 3% Tween 80.Standards (dissolved in 5% ethanol) were 4-hydroxy-biphenyl (30 yug/ml) and 2-hydroxybiphenyl (12 yu.g/ml) .Cofactor solution (0.2 ml) contained 1.25 yjunolNADP+, 13 j x m o l G-6-P and 1.5 units G-6-P dehydrogenas

o10 minute incubation at 37 C.

Incubations were stopped by the addition of 0.5 nil 2 0 % trichloroacetic acid. The incubation was extracted for 15 min with 7 nil n-heptane (containing 1 % (v/v) isoamyl alcohol) on a rotary shaker. A 2 ml aliquot of the n-heptane extract was extracted for 15 min with 7 ml 0.1M NaOH on a rotary shaker. The two phases were separated by centrifugation (2,000 rev./min, 10 min) and the n-heptane was sucked off. 2 ml of the NaOH extract was pipetted into a spectrofluorimeter cuvette and mixed with 0.5 ml 0.25M sodium succinate, and the fluorescence measured on a Perkin-Elmer spectrofluorimeter The excitation/emission wavelengths for 4-hydroxybi- phenyl and 2-hydroxybiphenyl were 2 7 5 / 3 3 8 nm and 295/415 nm respectively.

(ii) Ethylmorphine N-demethylation.A modification of the method of Holtzman et al.

(1968) was used. Ethylmorphine was supplied us a gift by May and Baker Ltd., Dagenham, England. Incubations

were set up as follows:

Test Blank Standard(volumes in ml)

Microsomes 0.2Buffer 0.5Cofactor solution 0.2lOOmM MgSO^ 0.1Standard -Substrate 0.3

Final volume 1.3 ml. * added after the incubation.

Microsomes were suspended at 10 mg protein/ml.Buffer was 0.3M Tris-HCl, pH 7-4.Substrate was 50mM ethylmorphine in distilled water.Standard was formaldehyde in the range 1-1.5 j x mol/ml

— 1 — 1(standardised by €• = 8,000 l.mol .cm for the HCHO-Nash reagent complex)Cofactor solution (0.2 ml) contained 1.25 LunolNADP+, 15 j x m o l G-6-P and 1.5 units G-6-P dehydrogenase.

O10 minute incubation at 37 C.

The reaction was stopped by the addition of 1 ml 15% ZnSO^ and 1 ml 2:1 freshly mixed saturated BatOH)^: saturated sodium tetraborate. After mixing, the precip­itated protein was centrifuged down. A 2 ml aliquot of the protein-free supernatant was mixed for 40 min at 37°C with 2 ml freshly prepared 4M ammonium acetate containing 0.4% (v/v) acetylacetone. The absorbance was measured at 412 nm.

0.2 0.20.5 0.20.2 0.20.1 0.1

0.3*0.3* 0.3*

(iii) Hexobarbital oxidation.An adaptation of the method of Gilbert and

Goldberg (1 9 6 5) was employed in this assay. The reaction was followed by measuring the disappearance of substrate. Incubations were as follows:

Test Blank Zero time (notincubated)

(volumes in ml)

Homogenate 2.0 2.0 2.0Buffer 0.2 0.7 0.2Cofactor solution 0.2 0.2 0.2lOOmM MgCl2 0.1 0.1 0.1Substrate 0.5 - 0.5

Final volume 3 ml.

Liver whole homogenate (33%) in 0.25M sucrose,0.1M phosphate, pH 7*4, 0.001M EDTA.Buffer 0.1M Na2HP0^..NaH2P0^ , pH 7-4.Substrate 20mM sodium hexobarbital.Cofactor solution (0.2 ml) contained 1 unol NADP+,3-75 jjunol G-6-P and 0.2 units G-6-P dehydrogenase.15 minute incubation at 37°C.

The reaction was stopped by saturating the medium with NaCl and extracting with 10 ml n-heptane (containing 1.5% isoamyl alcohol) and 1 ml M NaH^PO^ for 20 minutes.A 2 ml aliquot of the n-heptane was extracted byshaking with 6 ml 0.8M Na^PO^ (brought to pH 11 with10M NaOH) for a further 10 minutes. The absorbance of the aqueous phase was measured at 245 nm.

(iv) Nitroanisole O-demethylation.A modification of the method developed by Netter

and Seidel (1964) was used to determine this activity. Incubations were set up as follows:

Test Blank Standard(volumes in ml)

Microsomes 0.2 0.2 0.2Cofactor solution 0.2 0.2 0.2lOOmM MgSO^ 0.1 0.1 0.1Standard - - 0.2*Substrate 1.2 1.2* 1.0*

Final volume 1.7 ml. * added after the incubation.

Microsomes (10 mg protein/ml) were suspended in0.1M Na^HPO^.NaH^PO^, pH 7.8.Standard was 0.3mM p-nitrophenol.Substrate was 4mM nitroanisole dissolved in 0.1M phosphate buffer, pH 7 , 8 .

“f*Cofactor solution (0.2 ml) contained 1.25 yjunol NADP , 15 yuimol G—6 —P and 1.5 units G-6 -P dehydrogenase.15 minute incubation at 37°C.

The reaction was stopped with 0.5 ml 2 0 % TCA.The protein precipitate was centrifuged down and 1.4 ml of the supernatant was mixed with 1.5 ml 4M NaOH. The absorbance was measured at 4l0 nm.

(v) Acetanilide 4-hydroxylation.The method developed by Krisch and Staudinger

(l96l) for assay of this activity was modified. The

incubation was not stopped with acid as Krisch et al. (1964) found this lead to artefactual results. Incubations were as follows:

Test Blank Standard(volume in ml)

MicrosomesBuffer

0.30.9

0.3 0.9 0.2 0 . 1

0.30.40.20.10.5*0.5*

Cofactor solution 0.2lOOmM MgSO^ 0.1StandardSubstrate 0.5 0.5*

Final volume 2.0 ml. * added after incubation.

Microsomes were suspended at 10 mg protein/ml.Buffer 0.3M Tris-HCl, pH 7.4.Standard 0.4mM 4-hydroxyacetanilide.Substrate l6mM acetanilide.

+Cofactor solution (0.2 ml) contained 2 ^mol NADP ,15 yumol G-6-P and 1.5 units G-6-P dehydrogenase.

30 minute incubation at 37°C.

The incubation medium was saturated with solid NaCl and shaken with 10 ml diethyl ether (containing 1.5% isoamyl alcohol) for 5 minutes. 8 ml of the ether phase was sucked off and 1.5 ml of the aqueous phase was added to 1 ml Folin-Ciocalteu reagent (diluted 1:6 (v/v)with water). This was mixed, incubated at 37°C for 30 minutes and the absorbance measured at 691 nm.

(vi) Aniline 4-hydroxylation.The method of Nakanishi et al. (l97l) formed the

basis of this assay, the 4-aminophenol formed being measured by a method similar to that of Brodie and Axelrod (1948). Incubations were set up as follows:

Test Blank Standard(volume in ml)

Microsomes 0.3 0.3 0.3Buffer 0.7 0.7 0.6Cofactor solution 0.2 0.2 0.2lOOmM MgSO^ 0.1 0.1 0.1-Standard - - 0.1*Substrate 0.5 0.5* 0.5*

Final volume 1.8 ml. * added after the incubation.

Microsomes 10 mg protein/ml.Buffer 0.3M Tris-HCl, pH 7*4.Standard 0.5mM p-aminophenol in 0.01M HC1.Substrate 20mM aniline hydrochloride.Cofactor solution (0.2 ml) contained 2 jimol NADP ,15 yuumol G-6-P and 1.5 units G-6-P dehydrogenase.

' o15 minute incubation at 37 C.

The reaction was stopped by saturating the incubation with NaCl. This was extracted with 10 ml diethyl ether (peroxide-free) containing 1.5% (v/v) isoamyl alcohol, for 30 minutes on a rotary shaker. A 7 ml aliquot of the ether extract was shaken with 4 ml 1% phenol in 0.5M K^PO^ for 30 minutes, and was then left to stand for at least 60 minutes. The ether was

sucked off and the absorbance of the aqueous phase measured at 630 nm.

(vii) Thiabendazole 5-hydroxylation.The method used was developed by Wilson, Parke and

Cawthorne (unpublished work). The incubations were set up as follows:

Test Blank Standard(volume in ml)

*1 0 , 0 0 0 g* supernatant 0..5 0 . 5 0 . 5

Buffer 0.7 0.7 0.6Cofactor solution 0.2 0.2 0.2lOOmM MgCl2 0.1 0.1 0.1Standard - - 0.1*Substrate 0.5 0.5* 0.5*

Final volume 2 ml. * added after incubation.

Buffer 0.3M Tris-HCl, pH 7-4.Standard 5 -hydroxythiabendazole (l^imol/ml in 0.1MHC1) .Substrate lOmM thiabendazole (2-(41-thiazolyl)benzimidazole) in 1 % Tween 80.Cofactor solution (0.2 ml) contained 2 j j a n o l NADP+,10 yjjnol G-6-P and 2 units G-6 -P dehydrogenase.20 minute incubation at 37°C.

The reaction was stopped by the addition of 10 ml ethyl acetate and 3 ml 0.3M sodium phosphate, pH 6.0 and shaken for 15 minutes. A 5 ml aliquot of the ethyl acetate extract was shaken with 5 ml 0.1M HC1 for 15

minutes, and the HC1 extract (diluted 1 in 10 with 0.1M HCl) was assayed fluorimetrically at an excitation wavelength of 336 nm and an emission wavelength of 5 3 0 nm.

(viii) Zoxazolamine hydroxylation.The methods of Conney et al. (i9 6 0 ) and Juchau et

al. (1 9 6 5) were considerably modified to give thefollowing procedure for measuring the disappearance of zoxazolamine. The incubations were as follows:

Test Blank Zero time (not incubated)

(volume in ml)

*10,000 g' supernatant BufferCofactor solution 0.5

0.1

0.50.4

0.50.90.50.1

0.50.40 . 5

0.10 . 5

lOOmM MgSO^ Substrate 0.5

Final volume 2 ml

Buffer 0.1M Na^PO^ .Nat^PO^, pH 7-4.Substrate 300 xM zoxazolamine (dissolved in a very small amount of M HCl and diluted with water.

+Cofactor solution (0.5 ml) contained 1.25 Jtimol NADP 15 f t mol G-6-P and 1.5 units G-6-P dehydogenase.40 minute incubation at 37 °C.

The incubation was stopped with 0.5 ml M NaOH, and extracted by shaking with 10 ml n-heptane (containing 1.5% isoamyl alcohol) for 30 minutes. A 7 ml aliquot

of the organic phase was shaken with 2 ml 0.1M acetate buffer, pH 5*6, for 5 minutes. 5 ml of the n-heptane was then extracted with 3 ml 0.1M HCl, by shaking for 5 minutes, and the absorbance of the acid phase was measured at 278 nm against 0.1M HCl.

6 .3 « Results.

Spectral interactions.The binding spectra of five chlorofluoroalkanes:

CC13FCHC1 F

CC12F2

CC1 F-CC1F CC1F -CC1F

with microsomal preparations were examined. They were all of the type I class, exhibiting peaks at about 3 8 3 - 3 8 8 nm and troughs at about 420 nm. The spectra produced by a number of chlorinated methanes, ethanes and propanes were also type I. Of the compounds tested:

CC14CHCl^

CH2C12

CHClg-CH , CH2C1-CH2C1cciq-ch„, chci0-ch0ciJ J eL cLCHC12-CHC12 CC1„-CHC10

■ -CC1 -CC1

CC1 -CC10-CC1

only 1 ,2 -dichloroethane failed to give any sort of spectrally observable interaction.

Figure 6.1 shows a comparison of the spectra obtained with trichlorofluoromethane and carbon tetra­chloride. The spectra have been redrawn, for this figure, on a straight baseline. Such an ideal baseline was not obtainable directly on the spectrophotometer used, as provision for balancing the two light beams was made only every 40 nm.

Determination of K values.------------------ s-------Considerable problems were involved in the accurate

determination of K values for the chlorofluoroalkanes.sAll the compounds have vapour pressures close to or greater than 1 atm. Sequential additions to the sample cuvette, made by removing the stopper, involved the risk of loss of vapour and consequent repartitioning of substrate between the microsomal suspension and the' vapour phase. The use of a ‘Subaseal* septum in place of stoppers allowed repeated additions of substrate by injection through the septum. However, Figure 6.2

VD

0UftbO•Hft

Hs\ft

•H0-POftftb£s

LT\

(M

tofto•rltOft0fttoftto

rH0£otoofto•rl£' ft 0 > •rf H

-P•ftft

.ft

T30.-P•H.f t

rftK0ftft

-Po0ftto

fcJ!ft•HTJft•H.QT*0to•rlTS•rlKo

VD O CO CM CA<t<

0 «H ft TJ

VO O CO CM CA-cf

in

0 *rlft TJ

incn

Ao

compares the plots obtained for carbon tetrachloride using glass stoppers and 'Subaseal* septa. It is apparent that the Kg values obtained with the ‘Subaseal1

septum method are greater than those obtained by using glass-stoppered cuvettes. The explanation for this is that the septum, which is made of silicone rubber must have absorbed approx. 200 ig carbon tetrachloride before becoming saturated. The actual concentration in the cuvette was thus smaller than the calculated concentration, giving a smaller absorbance change than expected and resulting in the shift of the S/AE versus S plot upwards. A similar result was found with 1,2-dichloro-l,1,2,2-tetrafluoroethane. Even making single additions of substrate to fresh microsomal suspensions for each substrate concentration was not entirely satisfactory. Kg values for the more volatile compounds are therefore only to be regarded as approximate.

Using microsomal preparations from the rat, the following approximate IC values were determined in an initial series of experiments:

mMCC13F 3 - 2

C C 1 2F 2 6 -°CHC12F kOCC1F2-CC1F2 30

CCl^ 2 . 0

chci3 6 . 6

CH2C12 1-7

Figure 6.2.

Determination of the K for carbon tetrachloride in ---------------------- s-----------------------------microsomal preparations from the plienobarbital-pr etr eat ed

mouse, using glass and 'Subaseal1 stoppers.

+ 'Subaseal* stopperAE

Glassstopper

20

-1substrate

concentration(mM)

The determination of K was not affected by thes Jprotein concentration of the microsomal suspension (Figure 6.3) • Spectral constants were thus measured at microsomal protein concentrations in the range 1 - 3 mg/ml suspension.

The K plots frequently showed a break, thus yielding two intercepts on the x and y axes (Figures 6.2 and 6.4). The break was not reproducible, changing position with respect to substrate concentration. In cases not showing the break, the derived from the plot was generally the higher of the two possible values,i.e. the lower affinity.

Tables 6.1, 6.2 and 6.3 show the K valuessdetermined for various polyhalogenated alkanes with rat microsomal preparations, the effect of various pre­treatments of the animals on K values for trichloro-sfluoromethane and carbon tetrachloride, and the comparison of Kg values in various species.

Pretreatment of rats with phenobarbital and 3-methylcholanthrene altered the magnitude of the difference spectrum produced by 2mM trichlorofluoro­methane , measured at identical cytochrome V - k ^ O concentrations. The spectrum was greatly enhanced (1 0 0%) in microsomes from phenobarbital pretreated rats, but preparations from 3-niethylcholanth.r ene treated rats showed a'reduction ( k 0 % ) in the spectral interaction (see Figure 6 .l6 ). The values for trichlorofluoro­methane in control and phenobarbital pretreated rats

W m co voON

VOCD

3b0•Hfti

-P

cd-pwriooH■ cdu■poCDA05

CDfC5■p<hOfto•H

-Pcda•HsSh<D+>CD

CDrft-P

Oflo•H

■Pcd

■P

CDOtiooPi•HCD

-POUftHcdSowoho

•rlS<ho-poCD<H

•HI

•CD

•rluOi—IiftocdU■PCD-P

uo<H

t-C

CM

COo

o

01

l/substrate

conc

entr

atio

n

F i gure 6.4K_ plots for halogenated methanes with microsomal

preparations from the guinea-pig.

A E

150

CHC1

100

CC1

10

Substrate concentration S (mM)

Table 6.1

K_ values in rat microsomal preparations.

mM

CH2C1 2 10.25 + 3-75 (2 )CHC1 3 2.94 (2 )CCl^ 1.54 + 0 . 1 8 (2 )CC-1 F 3-19+0.15 (2)CH2C1 -CH2C1 no spectrumCHC1 2 -CH3 11 + 2 (2 )CHC12 -CH2C1 3-7 (l)CC1 3-CH3 4.6 (1 )CHC12-CHC12 2.78 + 0.12 (2)CC1 -CHC1 1.1^ (l)

3 2

CC1 -CC1 0.75 (1 )CC1 -CC1 -CC1 „ 0 . 3 8 (2 )J 2 jCC1 2F-CC1F2 5.1 (1 )CC1F2 -CC1F2 8 . 1 (1 )

Figures in parentheses show the number of estimations.

Table 6.2

Effect of pretreatment of male rats on the spectral constants for trichlorofluoromethane and carbon

tetrachloride.

(the number of animals used is given by the figures inparenthesis)

Pretreatment Kg (mM)cci3f CC14

Phenobarbital

Control 2.25 (2 ) 1.15 (3)Test 1.45 (2 ) 0.4l (3)

Control 1 . 3 8 (3)Test 0.49 (3)

Control 1 . 8 3 (3)Test 0 . 6 5 (3)

Trichlorofluoromethanei

Control 2 . 2 0 + 0 . 0 2 (3 ) 1 . 8 7 + 0 . 0 1 (3 )

Test 1.58 (3) 1.40 (3)

Carbon tetrachloride

Control 1.85 + 0.25 (2 ) 1 . 6 6 (3)Test 2 . 8 0 + 0 . 1 0 (2 ) 1 . 9 0 (3)

Control 1.34 (2 )

Test 1 .44 (3)

Table 6.3

Species differences in spectral constants.

K (mM)s

CC13F c c i 4 chci3

rat 2.7^ +0.49 (1 1) 1.39 + 0 . 0 7 (l8 ) -

mouse -HoCO .

•o

0 . 8 0 (l) 2.7 (1)

hamster 4.20 + 0 . 1 5 (3) 1 . 2 0 (3 ) -

guinea-pig 2.42 + 0.42 (2 ) 1 . 7 2 + 0.37 (2 ) 3.7 (1)

chicken 3.33 (l) - -

Figures in parentheses indicate the number of estimations.

were only slightly different (3-04 and 2.71 mMrespectively), but in 3-methylcholanthrene pretreatedanimals the K was increased (9.65 mM) . s

Effect of isooctane extraction.Extraction of microsomal preparations with

isooctane abolished the aerobic difference spectrum produced by trichlorofluoromethane, even at high concentrations. Some slight spectral change was observed but this was independent of the concentration of substrate added (Figure 6.5). At very high concentrations, a modified type II spectrum was observed with carbon tetrachloride. This is shown in Figure 6 .6 , compared with the spectrum produced by aniline, a type II substrate. The Kg value for aniline in microsomal preparations from male rats (2.6 x 10 mM) was not altered after isooctane extraction (2 . 5 x 10 ^mM).

Effect of phospholipase digestion.Digestion of the microsomal preparation with

phospholipase C also resulted in some loss of type I spectral interactions due to chlorofluoroalkanes and chlorinated hydrocarbons. • The spectra produced by 4mM trichlorofluoromethane with untreated, isooctane extracted and phospholipase C digested microsomal preparations are shown in Figure 6.17- In. the case of phospholipase C, the spectrum was altered, the peak increasing and broadening and the trough disappearing.

«AVO<Du3b O

• r lfa

V0•P0ccsPi-PKCD1CDPiccs

•poooto

• r l

"0PiCCSHOPi■P5 0 o£•p• r l

£

CDPiCCSA-P<D6

o u 02 H fa 0 u or-1Ao• HPi+>

faOtoPio

• r l-P

OCCSPiCD-PPi• H

+>'

LACA

O

CCSPi•PO0fa

Pi• r l

0bOPiccSAO

0oPiCCSrOPiotoA

CCS

OI oI

Figure 6.6Interactions of* CCl; ("". “) with control and isooctaneextracted microsomal preparations compared to the aniline

(------ ) spectrum.(Both preparations 2 mg protein/ml)

a

Control Isooctaneextracted

\/ ' i i i i i \

41.4mM CC1,.

4. 4mM aniline

nm350 430

\ / V /

390nm

4?0

bas elines)

Treatment with phospholipase D by the method described (6.2) did not affect type I spectral binding. Treatment of microsomal preparations with phospholipase D at pH 5*6, as. described by Eling and DiAugustine (1 9 7 1)» was found to result in approx. 9 5 % destruction of 'cytochrome P-450 (Burke M.D., unpublished observations).

Interference by chlorofluoroalkanes with type II spectra.Schenkman (1970) showed that the presence of low

concentrations (2 .5mM) of a type I substrate (hexobarbital) enhanced the type II difference spectrum of aniline.

Trichlorofluoromethane, however, interfered with the spectrum due to aniline. Figure 6.7 shows the effect of addition of 2mM trichlorofluoromethane on the spectra produced with 1, 2 and 4mM aniline.

Reduced spectra.The findings of Reiner and Uehleke (l97l) using carbon

tetrachloride and rabbit hepatic microsomal suspensions were confirmed using microsomal, suspensions from the untreated rat. Figure 6 . 8 shows the spectrum produced by carbon tetrachloride and Na^S^O^ compared to the aerobic difference spectrum. It was notedthat the spectrum altered with time and this is also shown in Figure 6 . 8 superimposed on the cytochrome P-450-C0 spectrum. Upon reduction of the microsomal suspension with Na^^O^ the type I difference spectrum was abolished and (a) was recorded after 15 seconds.

Figure 6.7The interference between trichlorofluoromethane andaniline difference spectra in microsomal preparations

from the rat.

Baseline is the heavier line. .

2mM CC10F

t I j nm350 430 510

aniline aniline + 2 mM CCl^F

2 mM

k mM

Figure 6.8Formation of the carbon tetrachloride reduced spectrum

with microsomal preparations from the rat.

E = 0.05

baseline

oxidised spectrum

reduced spectrum

(a) (b) (c) (d)

E =

cytochrome P-450 CCl^ reduced spectrum

This showed a peak at about 420 nm. The peak at 420 nm decreased over the next 5 minutes and a peak at 4^4 nm appeared (b-d).

Trichlorofluoromethane was found to interact in a very similar fashion, giving peaks at 422 nm and 452 nm. The magnitude of the spectrum was greater than that caused by carbon tetrachloride. Figure 6.9 shows reduced spectra of trichlorofluoromethane and carbon tetrachloride compared to the cytochrome P-450-C0 spectrum. The formation of the reduced spectrum of trichlorofluoromethane in rat microsomal suspensions was also time dependent. With trichlorofluoromethane and carbon tetrachloride (final concentration 2mM) the time required to achieve maximal spectral interaction was found to be 17-18 min and 5-5-6 min respectively (Figure 6.10). No difference was found in the rate of formation of these spectra due to the prder of addition of the substrate and the Na^S^O^ to the microsomal suspension. Incubation of trichlorofluoromethane with the microsomal preparation for 20 min prior to the addition of Na^S^O^ made no significant difference in.the rate of spectral interaction. The spectrum due to carbon tetrachloride, however, reached a maximum within about 3 min.

Three halogenated compounds were found to give reduced spectra with peaks at about 450 nm:

c c i3fccik

CC1 F-CC1FA t-i

452 nm 454 nm 452 nm

F i gure 6.9Reduced spectra due to trichlorofluoromethane andcarbon tetrachloride compared to the cytochrome P-450-C0

spectrum.

0 . 2

. 1

510390 iS* nm//4 5 0

ll ^

CO-0 . 1

CC1 _F

CC1.

Rate

of formation

of reduced

spectra

in microsomal

suspensions

from

the rat

facaHO

CMVOCO

< o^ IAICM

LA

Figure

6-10

These are shown in Figure 6.11 together with the spectrum produced by pentachloroethane (peak at approx. 420 nm). This latter spectrum was typical of the spectra produced by the other chlorofluoroalkanes and chlorinated hydrocarbons tested. Table 6.4 gives the maxima of the reduced spectra obtained with these compounds and micro­somal preparations from the male rat. Similar results were found with microsomal preparations from rat, mouse, hamster, guinea-pig and chicken.

Halothane (2-bromo-2-chloro-l,1,1-trifluoroethane) also yielded a reduced spectrum in chemically reduced microsomal preparations. The peak was smaller than that produced by carbon tetrachloride, with its maximum at 471 nm, and was formed within 10 minutes (Figure 6.12).

The rate of formation of these reduced spectra wasfurther investigated with a range of concentrations oftrichlorofluoromethane (Figure 6.13)• It was found thatthe initial rate of formation was dependent on theconcentration of added substrate. The curves obtainedwere reproducible (Figure 6.l4). A plot of thereciprocal of the initial rate of formation againstreciprocal substrate concentration yielded a Lineweaver-Burk type plot (Figure 6.13)- The reciprocal of theintercept on the substrate (x) axis gave a value of 1 . 3 3

mM for the "spectral K ".mIn vivo pretreatment of the rats and in vitro

treatment of the microsomal preparations affected the

Table 6.4Maxima of spectral interactions in Na^S^O^-reduced

microsomal suspensions of the male rat.

Compound Final concentration Wavelength of(mM ) spectral maxima (nm)

c c i3f 0.5 422 and 452

CC1 2 F 2 3-5 -CHC1 F 2 . -

CC14 1 4l8 and 454CHC1 10 420 small

1 O

i W to o H to 10 420 v. small

CH2C1-CH2C1 20 422 v. smallCHC12 -CH3 20 -

cci3-ch3 10 422CHC12-CH2C1 10 420CHC12-CHC12 10 420CC1 -CHC10J 2 2.5 4l8-420 smallCC13-CC13 3 420 small

CC12F-CC1F2 5 approx. 420 and 452CC1F2-CC1F2 1 -

CF^CHBrCl 5 approx. 426 and 471

Figure 6.12The reduced spectrum of halothane in rat microsomal

preparations.

A1▼

430 3 1 0 nm

0.025A E

(471-510 nm) 0 . 0 2 0

0.015

0 . 0 1 0

0.005

105minutes

<H

PJctf•P0

VOcn W

O OCvMrH O O

CO

IA in©c

© o,VO

•CO

O] o CO VO oo

ow o <J in03in

Figure

6.13

(452-500

in

a s

ma•p§•Hs

rHVOo3bO•Hfa

Oo

oo

Reproducibility

of rate

of reduced

binding

of tr

ichl

orof

luor

omet

hane

.

V .

5 00

4oo

300

200

Spectral K 1 . 3 3 niM'100

105 _ |l/substrate concn. (mM )

Figure 6 .15

trichlorofluoromethane reduced spectra only to a small extent.

The small peak at about 420 nm was abolished in preparations from phenobarbital pretreated animals, but the rate of formation of the 4.52 nm peak and the maximal absorbance change were not altered. 3-Methylcholanthrenepretreatment of the animals resulted in the formation of smaller reduced spectra (Figure 6 .l6 ) at a slower rate (0.0085 units E/min compared to O.O182 units E/min in phenobarbital treated and control animals at a final substrate concentration of 2mM with 2.5 nmoles cytochrome P-450/ml).

Isooctane extraction and phospholipase C digestion of microsomal preparations did not abolish the reduced spectra. Both treatments caused increases in the magnitude of the 422 nm peak, although only isooctane extraction increased the magnitude of the 452 nm peak of the trichlorofluoromethane spectrum (Figure 6.17)*The rate of formation of the reduced spectrum was only affected byjphospholipase! C. digestion of the

microsomal preparation (Figure 6 .l8 ) , which approximately halved the initial rate. Phospholipase D treatment failed to have any effect.

Aniline interfered with the reduced spectral interactions to a small extent:

F i gure 6 . 16Effects of pretreatment of rats on the oxidised and

reduced spectra of trichlorofluoromethane

Oxidisedspectrum

ControlP-450nmol/ml

Phenobarbital pretreated

P-450 1 .7 nmol/ml

Reduced

3-Methyl- cholanthrene pretreated

P-450 1.7 nmol/ml

spectrum

!>• T—I •VO

0upibO

•rlft

T50O0Pi

*dpid

TJ0to•H

•rlXO0£+>PioT5•Hft•rlHOAfttooAftHd£otooPio

•rls0£

-p

ftOPi0

•rl■POdPi•PK0ftO■Po0ftft

65)

0todft

•rlHO£ft0O£ft

0PidrCj-P0soPiOPir—IftOPiOH£o•rlu-p

ftOsPiPi-PO0ftto

Pid

-pooo0

H

HOPi•PPiOO

OrHin

ocnch

ocn

oincn

Cytochrome P-

50

adjusted

to 1.

1 nmo

Rate

of formation

of the

CC

l^F

reduced

spectrum

in extracted

microsomal pr

epar

atio

ns.

(cytochrome

P-^5

0 adjusted

to 1.

1 nm

ol/m

l)

73 730 0

•P +>to to0 0

73 b O 600 •H •H

■P 73 73Oa O Qu

•p 0 0k to to0 d dft ftCD •rl •HPi H H

H d 0 00 -P rd -du O ft ft

-p 0 to toPi 0 0 o0 to rd •do H ft ft© o < <

oCO

oVO

o

OCM

W<

o0m1CMin•4«

minut e s

Initial rate of AEmax.formation of the (452-CCl^F 452 nm peak 500 nm)

(AE/min)

Control 0.0184 0.111+ ImM aniline 0.0172 0.103+ 2mM aniline O.Ol49 0.095+ 4mM aniline 0.0108 0.094

Effect on measurement of cytochromes b _ and P-450«Neither trichlorofluoromethane nor carbon tetra­

chloride had any effect on the measurement of cytochrome . Both compounds, however, reduced the magnitude of

the cytochrome P-450 spectrum, e.g. by 20% at a final concentration of 2mM. At a final concentration of lOmM, .carbon tetrachloride caused a 40% reduction in the cytochrome P-450 spectrum. Even bubbling the microsomal suspension for several minutes with CO did not restore the cytochrome P-450 spectrum to the amplitude of control measurements.

Destruction of cytochrome P-450 by incubation at 37°C.Microsomal suspensions were stored at 0°C for at

least 30 minutes without destruction of the cytochrome P-450. Incubation at 37°C, either static or shaken (lOO cycles/min), caused destruction of cytochrome P-450. Without NADPH, 10-15% was destroyed in 30 minutes, whereas inclusion of an NADPH generating system in the incubation increased the loss of cytochrome P-450 to

40-50%, cts shown below:

Generating Cytochrome %,system P-450 remaining

(nmol/ml)

Zero time - 2.26 (2)30 min in ice - 2 . 2 0 (2 ) 98

30 min in ice + 2 . 3 1 (2 ) 102

30 min at 3 7 °C static 1 .90 (3 ) 8430 min at 3 7°C static + 1 . 1 5 (2 ) 51

30 min at 3 7 °C shaken - 2 . 0 0 (3 ) 89

30 min at 3 7°C shaken + I . 3 8 (2 ) 6 l

(figures in parentheses show number of incubations)-

The reported conversion of cytochrome P-450 to P-420 by incubation with carbon tetrachloride (Heni, 1971) could not be repeated. Both carbon tetrachloride and trichlorofluoromethane caused approx. 10% conversion of cytochrome P-450 to a spectrally unobservable product.

The presence of a substrate (zoxazolamine) protected the cytochrome P-450 during a 30 min incubation, but did not affect the interference by trichlorofluoro­methane or carbon tetrachloride:

cytochrome P-450 AE (450 - 500nm)expressed as % of zero time value.

Blank (no substrate)+ substrate (zoxazolamine)+ substrate + 10 j f L CCl^+ substrate + 10 j x 1 CCl^F

+ substrate + 10 j x l C^Cl^F^

100%

79%

5 2%2 k %

7 9 %

It can be seen that 1 ,1 ,2-trichloro-1 ,2,2 -trifluoroethane prevented the protection of cytochrome P-450 by substrate, but did not further decrease the concentration of the cytochrome.

Effects on hexobarbital sleeping times and zoxazolamine paralysis times.

\Tables 6 . 5 and 6 . 6 show the effects of chloro-

fluoroalkanes on the. hexobarbital sleeping time of female rats. Trichlorofluoromethane (lg/kg) significantly increased the sleeping time in two experiments.Diehlorofluoromethane (lg/kg) showed no significant effect, whereas 1 ,1 ,2-trichloro-1 ,2 ,2-trifluoroethane (lml/kg) decreased the sleeping time significantly.

Zoxazolamine paralysis times were not affected by trichlorofluoromethane (lg/kg)(Table 6 . 7 ) •

In vitro enzyme assays.The in vitro metabolism of the four type I binding

substrates (biphenyl, ethylmorphine, hexobarbital and nitroanisole) was not affected to any significant extent

LT\VO<DH

42d

Hd+>•rl42Ud42 OX043043-PPiOft43faO•rl0£inT3O

42

b OX\bOinO*3sdbO43 \bO

0ftd43-P0B.ouo3rHftOuorH43o•HPiftftOfto0ftft0043H:

toft(3PioS3•rl42HdPidftto•Hts0iHdB0ftftO0£•HftbOPi•rlft00rHto

rs0 rH inu 0 rH rHd U • CM o Hf t ft f t O i o d£ P3 i • • • U0 0 JZi o o 0o ov_ n0 0ft f t too •

to0cn 0 K/—\ CM • CO IN. 0bO s3 • CM • • • d ‘ HS3 •H cn ft n- cn CM 43 • f t•rl £ f t • 0 0ft + 1 + 1 + i + 1 + 1 0 ft Pi0 £ d0 0 ft n - in o O 0 0 bOH £ • • • • • u H Pit/3 •rl n- VO ON cn in 0 0 •Hft VO in CO CM cn 2 ftrH rH 43ft in bOTD X—s 0 43 •H •0 rH P ft u toSh 0 0 0 0d u rH ft Xft ft • ft ft 43 TJ 0 0£ Pi . ft o o o 0- rH0 0 1 • o I o •H > to fto o £ • Pi . •H ■ to 0

o o ft 0 0 u0 O rHft ft 0 bO

Pi T5 Pi• Pi •Hft to d ft✓-\ in • H 43

Pi 53 CM rH in CM •H 0 rH bO0 •H • • •' • • d •H•rl £ O O o o o ft ft Pift 0 Pi •HO + 1 + 1 + 1 + i + 1 f t 0 42 ft3 0 O o Pi OTJ £ ON VO VO ON o 0 dP3 •H t • t • •n> 42 53H ft CM CM rH in Pi 0 •H•H • X drH 0 bO/—s d 43 0ft ft Pi43 •H ft

bO 42 0 TJ/—s n- cn •H U S3bO • • 0 d Pi dON VO -a* VO CM % 42 O• • • • 0 •H to+> cn CM rH + 1 + 1 in X ft to>(«■* T3 0 O o

+ i + 1 + i m O . o 43 0 rHin • • 42 •rjX3 in cn ft cn ON 0 Pi Pi0 VO in. n- cn CM bO ft •H 0ft ft ft X 0

\ Pi S3 £bO o 0 4J£ •H 0 0

Pi £ 42ft to O ft ftO rH O 0 rH

d VO ON VO is. O ft 53 42 d• S ft >—' •H >0 •H £ 0 U£ Pi rH £ 0

d d O •H ftft ft ft Pi•H •H42 0 • •

bO bO bO Pi ft 0 • •43 X X d d £ 0\ \ \ 42 0 •H Eft fa fa fa 0 H ft •H

a cn cn cn X 0 ft0 rH rH rH 0 S3£ O O o 43 H 0 bOft H o o rH o •H Pid 0 0 £ 43 ft •H0 U bO bO u bO '3 f t O ftPi ft f t •H 0 3 0ft Pi in O P3 O T3 TJ 0

0 • * 0 • 0 53 Pi rHo o rH o rH 1/3 •H H C/3

1•CO

0 -pu cd0 PipirH 0ft Pi•H •HPi 42ft H1 cdOJ

PiCM cdr- ftrH CO1 •H0 £u0 0H H43 CCSO £■H 0u ftft1 ftCM 0rH 0■ £rH •rlXPi

-Pto

<d P3f— V •H

ft-p 0A 0to ip

VD -H CO• CDVD £ H

<D Cd-pH X •rl& 0 42cd 42 PiEh Ctf

to 42A 0\ Xto 0tH 43

0CD

43ftPi<d Pi

43 0-PCD f—ss -p0 43u to0 •rlPi 0i—I >ft0Pi *00 0

H 4243o to•rl MTJ \ft

H£

0

-PrH v_<<

OCD 0ft Pift cdCD 43

0 -p043H

X0 HPi 0

Pi • ftft -P to o£ Pi I • o rH0 0 £ • cdo O o ft •

•H CO0 rO 0ft ft U Kcd 0rQ rH-•— s 0 ft

Pi n- V D K 0.to •rl • • • 0 PiP3 £ CM w CM 41•rl v_ ' toft + 1 + i + 1 ft Pi0 0 0 •H0 £ CM cn CO ftH •H • • • P3 41C/2 ft V D r- CO 0 tocn cn CM •rl •Hft Pio0 ft

■r~5 0 •'d Pi CO0 H •rH CO 0Pi 0 CO KccS Pi O • 0 0 0ft ft ft to Pi H rH

' £ Pi 1 • • • 0 ft0 0 o •23 • ft t TS 0o o • 0 0 Pi PiN-/ 42 ft cdo cd toft ft Pi 0 rH PI­•H rH cd TH

£ 0 ft ft✓-N •rl 43P3 Pi rH V D m o rH 4! to0 •rl • • • c n Pi •rl•H £ ft O o 43 cd Pi-P v_ ' 0 ft 420 + 1 + 1 + i ft <12 O ftpi 0 cd K 0X £ Cn Os o 0 •d 0Pi •rl • • t rH 0 43 PiH ft I'- i n r- O > •rl

•rl ft cdH 0 0 toO 041 0 P3 Pi■ ft Ph 00 •rl X-P CO ft P343 cn co cn Pi rH O cdto • • • •rl 0 0

•rl . in n- CM Pi *r-3 CO0 ft Pi CO£ rs + i + i + 1 H Pi •rl 0to .rH 0 rH—' o r- ft Cd o PiX • • • Pi 0 P30 CO f t O 0 0 0PQ cn £ 0

ft ft f t X • ft £0 ft 0 fttfi o 0fiu0 •rl 42X rH

m /— s cC Hft rH U1 ft > cd0 cd cd 43 u >£ ft r- O N £ to 0 U• *H ft •rl ft 00 p3 0 0 Pi fta cd Pi £ •rl P3cd •rl

X >> • •H X 0 • •cd 0 £ 00 42 •rl £CM Pi ft •H

ft 0 to ft-p H pi A PiPi O H \ 0 to0 H O ft to •H Pi£ 0 ft 1 0 £ ft •rl-p Pi CM ft Pi O Picd ft H CM 0 o Pi 00 Pi O H rH o X 0Pi 0 HI O 43 rH Pi rHEh o o O O >_/ H to

0a

•rl-P

0•rl0|A

H0d0.ft0d

•rl£0

HOSI0«oSI

0A-p

do+>dbQ•HI'- 0£

VD

0 73H 0,Q ,Q

• £-1 b€d \bO

0d0d-P0aouodHft0do

Hdo•rld

•P

ft0

■Po0ftft00Aft

73 r \0 Hd 00 d'f t ■P m£ d • t rH0 0 m in 00 o i « 1 • d

V -/ 53 53 •P0 dft -p 0o

•0 •d 0

•H 0/*—s rH CO £ K

0 d • • 0 0•rl •rl VD co rH pH0 £ CO 03 0 • ft>■> + 1 + 1 rH rH SI ft 0

H 0 • u0 0 rH CO + 1 + 1 K •rld £ • « o ft0 •H 03 o VD LA SI 73 0ft •P VD a- VD • 0

rH LA 0 •P 0VO -p O mrH 0 o

u rH0 d•rl •rH 73 ■u d •ft /—\ 0 0

-p 0d d 0 X

•rl bO d 0d d rH rH rH rH £ •rl •rl rH0 •rl • • • • 0 £ ft

•rl £ o o o o O £ 0 0-P rH rH dO + 1 + 1 + 1 + 1 !A 0d 0 0 73 SI f t73 £ A - VD CO CO -P 0 0 0d •rl • • • • 0 A

H •P rH rH 03 03 0 0 drH bO SI •H0 A 0\ f t bO

rH bO o 0£ dd

-p o 0 730 A - •rl d

-P 0d O

• 03 03 a - VD •rl 0 0 to­+> • • • • d *T> co£ 03 rH CO LA {A •H d 0

/—\ rH £ •rl rH5a bO + 1 + i + i + 1 H 0

73 w 0 rH d d0 - f t o to -A - d 0 0 0

PQ VD 0 - CO 03 o SI 0 0rH rH 0 £ £

Ti K -p -p0 0 0 00 N A Ao

73 rH rH0 0

0 0 • 5> >f t H d 0 u d0 cd 0 ■P 0 0

£ d 0 •p •P• •rl Os a - LA LA -p 0 d d

0 d 0 rH •rl •rl53 cd £ 0

o • • • •d H 0 00 5a £ £3 £ •H •rl

H •p -P -Pft 0-P 0 d 0d u 73 0 •rl0 rH H 0 0 •rl 0. £ 0 0 rH 5> -P rO

•P d ft d f t rd •rH O r—10 -P CO •P co O 0 d 00 d H d H •H 0 73 dd 0 O 0 O d 0 d 0ft o O o O ft u H ft

by the addition of trichlorofluoromethane (1 0 j x l ) to the incubation medium (Table 6 .8 ). Addition of carbon tetrachloride inhibited metabolism, except in the case of hexobarbital. However, the assay for measuring the disappearance of hexobarbital from the incubation medium was found to be rather poor and no significance can thus be read into the results.

The metabolism of the four type II binding substrate (acetanilide, aniline, thiabendazole and zoxazolamine) was significantly inhibited by carbon tetrachloride. Trichlorofluoromethane and 1,1,2-trichloro-1,2,2- trifluoroethane also inhibited the hydroxylation of zoxazolamine to a significant extent (Table 6.12). Thiabendazole 5-hydroxylation was not affected by either of these compounds (Table 6.11). Aniline and acetanilid 4-hydroxylations were however increased by the addition of trichlorofluoromethane to the incubation media (Tables 6.9 and 6.10). This did not appear to be a solvent effect as chloroform and 1 ,1 ,1-trichloroethane had no effect on aniline 4-hydroxylation (Table 6.10).

6.4. Summary. ,The chlorofluoroalkanes were found to interact

with microsomal preparations, yielding type I bindingspectra, and spectral constants (Kg) "the order of

- 2 -35 x 10 . to 10 M. Modification of the microsomal phospholipid by extraction or enzyme digestion abolished these aerobic difference spectra.

1A

+ 1+ 1 VD +1 CO

ONCOON

rH

rHOu-pPiooftotoCDb OCCS

•PPiCDO '

UCDft

VDON+ 1

VD

•OCDWtoCDUftXCD CO

ON

o

ON•

VDCDrHftccsft

-pCCSd0HccSfi0ftftd•Hdo■H-PccSrH

K0 u *0{Aft1-3*0

d♦HH*HdCCS

-p0o<

d•Hs\d•H0

- pouftbO£\

d0£d0ft0

d

rH•HdccS-P0oCCS>NK0 d

d

ft1-chto0HO

ftO

H0-P•HftdCCSftOd0ftft

fto

d0

- pdft

HOd-Pdoo

-pCCSd

d0-pCCS0dft0dft

rHOdftdoo

f t rH VD O LA ON VD CO oCCS ft O CA o LA ON ON0 ccS • • • • • • • •d d rH rH rH o o o o o

CO O CMCO CM VDtH tH

IA O <t< c a OrH rH

Oft

LA CA CA O rH rH

rH CA ON rH VD oA - A - CO O CA O o

• • • • • • • - •rH CM rH o rH rH tH o

CA VD A -VD VD VDrH rH

ftoLA

ft!CM

ftOLAftCM

ftOLAftCMO

Od•Hdo*HftdrH0to

orH

od•Hdo•HftdrH010orH

orH

ftoLAftCMO

d•Hdo•HftdrHoto

orH

oCM

farH

rH

+CMfa

rHrH o o o o

rH V -/ o rH rH rH o o0 1 '—' 0 1 rHd fa fa d fa faft CA CM f t CA CMd H H H d rH rH rH0 O O O O o O oo O O O o o o o

Table 6.10

Aniline 4— hydroxylation in the male rat.

Control levels are expressed as nmol 4— aminophenol formed/mg protein/min. Other levels are given as a percentage of the control levels.

Control 0.432 0 . 6 0 2 0 . 4 2 6 0 . 6 0 8

+ 240 nl CCl^ 72 81 80 -+ 10 j x l CCl^ - - - 85

+ 1 yjO. CC13F - 155, - -+ 2 j j J l CCl^F - 104 - -+ 5 yxl CC13F 111 154 -+ 10 j j J l CC13F 164 132 - 116

+ 50 ytxl CC13F - 204 - -+ 10 yul CC12F-CC1F2 - - 126 107+ io j x i chci3 - - 100 -+ 10 yjO. CC1 3 -CH3 - - 104 -+ 10 a1 cci^-chci2 - - 119 -

Thiabendazole

5-hy

drox

ylat

ion.

tootod

0d0*bO

IA •

CA

+ 1 ON rHto

-pcddod

•rlrOrHcddcdPto•H£0Hcds

/—siHnO 00 d p Pd P o • O •cd d 1 o to o toPM ft 0 • • * •B o o d o doo 0>_■< p

iH0 0<H d d0 P H 1 CM VD A- CAd cd VD ON LA ON0 >o

•d•H0 £K rH ’O O d P VD a- O VD1 SI 0 • • • • •LA cd > (M CM CM CA CMno •Hto d iH + 1 + 1 + 1 + ! + 10 0rH & to CO ON ON ON LA0 cd \ • • • •s •H *0 o o CO CO A-d ,3 0 LA CA CMP £d0<H

0<H H0 cdS VO VD VD VD VD• •H0 dd cd-

t—\

OP\XrH

opH OC ft Po CAO P rHP O CAO ftH ft\ CA0 cF CA H

d H H H OP O O o CMd O O o O0o + + + +

TOPiCd0•a•HdoHrdoCddp0pdorQdcdo<H. otO- po0

00£- pd00£- p0,Q0od0d0<H•rlTSPdcdo•rl•rldbO•rl•tQOd

tocd&0d0£SH

p•rl>•rlPOcd0rdPdo0ddpK•Hg

0•rldoH£ocddp0Pdo&dcdo\0dcd£p0£odOdrH<HOdoH£o•HdP0rdP

•H+>•H> to•H +>-P cOO dcO

0CD dto •rl

CM <0 &rH rH H• cOVO X

0 dCD u cOH d -PrQ tocO •rl

CD £d CD•rl rH£ cOcO BrH0 CAtQCO dK •rl0 >w/S3

P (c

ompared

to co

ntrol)

1

0.0

02

0.0

2

0.0

1

•iS• LA CA CO CA• * • •• • O o o rHd i n•rl + 1 + 1 + 1 + 1B + i\ o CO CA ONd d t • • •CD 0 o VO A-t> 0 rH•rl srHbOXd •0£d 'to IA LAd o • t • •0 .o VD IN- CAo rHod•rHsCOrH0tQcO •H VO LAX CQ • • . •0 ON S- CO [N-tQrH0Bd

CM CO A-< • • « •O VO LA CArH

d0 /—\•H d■P 0cO •rl,Q -Pd •rlo d CAd d Cd•rl cO fo CA

CA rH0 0 rH o rH-P d o O] Ov-/ o O od0 rH rH H H•rl-P 0

d•rl ■P o o Od d rH rH rHd 0C o + + +

clo•H-Pco

*§0 d•rl\HOK1CDd•HscOrHOS3cOXotQ

d d . d0 0 O H•H •H •H o-P -P ■P s(0 cO cO drQ rQ &d d d oo o o LAd d d • H•H •H •H |d d d

1•H •rl •H 0e s s -PcOo o O d

-=f CF -VO -ptoi 1 1 rQd< CQ o to

Trichlorofluoromethane, 1,1,2-trichloro-1,2,2-trifluoroethane and carbon tetrachloride formed spectrasimilar to the cytochrome P-A^O-CO spectrum in Na^S^O^-reduced microsomal preparations. This spectral changewas time dependent and was not abolished by extractionor digestion of the microsomal phospholipid. A'spectral K 1 was derived for trichlorofluoromethane, mand was found to be approximately 1 .3 3 mM, compared tothe spectral constant (K ) of 2.7^ mM.s .

Both types of spectral interaction were shown to occur in microsomal suspensions obtained from the rat, mouse, hamster, guinea-pig and chicken.

Hexobarbital sleeping times were increased by an oral dose of lg trichlorofluoromethane/kg body weight 10 min before the hexobarbital injection.

Type I substrate metabolism was not inhibited by trichlorofluoromethane. Metabolism of type II substrate was increased (aniline and aeetanilide hydroxylations) , unaffected (thiabendazole hydroxylation) or inhibited (zoxazolamine hydroxylation) by addition of trichloro- fluoromethane in vitro.

CHAPTER 7

DISCUSSION.

Distribution.The aerosol propellant chlorofluoroalkanes have

been reported to be of* very low toxicity (see 1.4) after both acute and chronic exposure. However, little work has been published on the absorption, distribution or excretion of these compounds. Dollery et al. (1970)showed absorption of chlorofluoroalkanes after inhalation of aerosol propellants by humans, but made the false assumption that, because the g.I.e. peak for 1 ,2 -dichloro-1 .1 .2 .2-tetrafluoroethane was only 10% of the height of the g.l.c. peak of trichlorofluoromethane, little1 .2 -dichloro-1 ,1 ,2 ,2 -tetrafluoroethane was absorbed. However, as shown in Chapter 2 (2.6), detection of trichlorofluoromethane by electron-capture techniques is about 100 times more sensitive than the detection of1 .2-dichloro-1 ,1 ,2,2-tetrafluoroethane. This would suggest that, in the studies of Dollery e t al. , approximately 10 times more 1 ,2-dichloro-l,1 ,2 ,2-tetra- fluoroethane was absorbed than trichlorofluoromethane.This would bring the blood level of the former compound considerably closer to the concentrations of chloro- fluoroalkane which were found by Jack (1971) to cause cardiac sensitisation to adrenaline.

The distribution studies in Chapter 3 were conducted after oral dosing of rats with chlorofluoro- alkane dissolved in ethyl oleate. The results cannot therefore be used to predict tissue levels after inhalation of chlorofluoroalkanes. However, the

results can give an indication of the levels that might be expected, as well as showing tissues with particular affinity for chlorofluoroalkanes.

1 ,2-Dichloro-1,1,2,2-tetrafluoroethane (560 mg/kg body weight) was absorbed slowly from the gastro­intestinal tract, the maximum blood level being reached 60 minutes after dosing (Table 3*2). The liver concentration closely paralleled the blood level, and the liver to blood ratio remained fairly constant (about 70) between 20 minutes and 3 hours after dosing. This suggests that the concentration found in the liver is the result of a passive partitioning process of the chlorofluoroalkane between lipids or binding proteins of liver and blood. The high lipid solubility of the chlorofluoroalkanes and the very high ratio of •concentrations in adipose tissue to blood achieved 4 hours after dosing suggest that the areas of high affinity for these compounds are lipid in nature. The concentration of 1 ,2-dichloro-l,1 ,2 ,2-tetrafluoroethane in lung tissue reached a maximum at 30 minutes, although there was no significant difference (P>0.10) between the concentrations at 30 minutes and at 60 minutes.These observations indicated that pulmonary excretion achieved its maximum rate at about 30 minutes and that lung uptake and excretion occurred at the same rate at approximately 60 minutes after dosing. 1 ,2-dichloro-1 ,1 ,2 ,2-tetrafluoroethane was shown to cross the blood- brain barrier, but the concentrations in both brain

and cardiac muscle were low (approx. 2 . 5 fAg/g* representing about 0.002% of the dose). The occurrence of a maximum in the level of 1 ,2-dichloro-1 ,1 ,2 ,2 - tetrafluoroethane in the adipose tissue only after 4 hours was probably a reflection of the poor vascularity of the tissue as well as the repartitioning of compound from other tissues. Although highly lipid soluble, the compound was rapidly cleared from the tissues and only traces (2 . 5 /xg/g) were found in adipose tissue after 19 hours. The half-life■in blood was approximately 50 minutes.

In distribution studies with trichlorofluoromethane ( 5 0 mg/kg body weight), the scatter of values at particular time points was reduced a little by starving the rats for 18 hours prior to the experiment. This suggested that absorption of trichlorofluoromethane, and possibly the other chlorofluoroalkanes, by the gastric route was dependent on the amount of material present in the gastro-intestinal tract. In male rats, the liver showed a maximum concentration of trichlorofluoromethane within 5 minutes, which was maintained for at least a further 25 minutes. The blood level was still rising at 15 minutes, after which time, unfortunately, no further estimations were made. In female rats, however, the blood and liver concentration curves each showed two similar peaks, at 7 minutes and at about 45 - 60 minutes. There are several possible explanations for this double peak. Firstly, some vapour might have escaped from the stomach during oral dosing to be absorbed through the

buccal and bronchial mucosae. However, after dosing a rat with a solution of chlorofluoroalkane in ethyl oleat the stomach was considerably distended by the volatilise compound for a period of up to 20 - 30 minutes, suggesting that the oesophageal opening was closed. Secondly, accidental introduction of some material into the pulmonary tract could account for rapid absorption, but as this double peak was consistently observed, and apparently only in females, accidental dosage into the lungs can be discounted. Thirdly, enterohepatic recirculation, resulting from biliary excretion of the compound, could give blood level curves with two maxima, as in the case of carbenoxolone (Downer et al., 1 9 7 0)? but biliary excretion of trichlorofluoromethane was considered to be most unlikely. It is thus difficult to find a reasonable explanation for this phenomenon without assuming that the gastric absorption of trichlorofluoromethane is biphasic.

Measurement of the concentration of trichloro- fluoromethane in consecutive samples of blood from each of six female rats gave six curves, five of which showed a double peak. The first peak was somewhat later than previously observed, but the second peak again occurred at about 45 - 60 minutes. A plot of the mean values of these curves yielded a much flatter plot (Figure 3*9)• From this curve, the blood half- life was estimated to be about 60 minutes, which was considerably longer than that reported in human studies (0.3 - 1.5 minutes) by Paterson _et al. (l97l)«

The difference between male and female rats in adipose tissue levels of trichlorofluoromethane was probably due to differences in vascularity. The epididymal fat pad of the male rat receives a better blood supply than the supra-renal fat in the female.

Excretion.Excretion of trichlorofluoromethane occurred

primarily through the lungs. In three experiments,92 + of the dose was exhaled unchanged within 6 hours. This was very much more rapid than the pulmonary excretion of unchanged carbon tetrachloride, which Paul and Rubinstein (1 9 6 3) reported took 18 hours for 80% excretion.

At six hours the total trichlorofluoromethane retained in the tissues amounted to much less than 1% of the dose, thus leaving approximately 4-7% unaccounted for. A small percentage of this may have been present in the tissues not examined, e.g. muscle. Excretion in the bile or urine would most probably require some metabolism to occur in order to make the compounds water soluble. As reviewed in Chapter 1 (l.7)» some halogenated compounds are metabolised and are excreted as glutathione conjugates or as mercapturic acids. However, it was considered that the detection of sulphydryl compounds or increased halide ion excretion in the urine at the concentration expected, i.e. 5% dose or less, would not be possible using

standard chemical techniques or methods such as ion- specific electrode coulometry. In order to detect metabolism in vivo, radioactively-labelled chlorofluoro­alkanes of high activity would be required. As these were not available, the in vivo metabolism was not studied.

Metabolism.Attempts at detecting in vitro metabolism of tri­

chlorof luoromethane gave negative results. No attempt was made to detect metabolism of the other chlorofluoro­alkanes because of their greater volatility and their greater content of fluorine atoms. Dichlorofluoromethane,which is analogous to chloroform, was not studied because of the difficulties in trapping the highly volatile possible metabolites and in obtaining adequate separation in the g.l.c. assay. With trichlorofluoro­methane , the most probable metabolite was considered to be the reductive dechlorination product, dichlorofluoro­methane . This latter was detectable when 1-10 j x X of a mixture of trichlorofluoromethane (9 9 -8%) and dichloro- fluoromethane (0 .2%) was added to a liver microsomal system (3 ml) in vitro. However, following incubations of trichlorofluoromethane with liver microsomes, no dichlorofluoromethane could be detected. This result was obtained with all the species tested, including phenobarbital-induced mice, which show one of the highest rates of general metabolism. Thus the extent

to which any in vitro metabolism of trichlorofluoromethane occurred was certainly less than 0 .2%, and possibly less

than 0 .0 5%.Results of attempts to detect free-radical formation

from trichlorofluoromethane by measurement of its effects on lipid peroxidation were equivocal. In one experiment increases in peroxidation were noted, in other experiments the principal effect was depression of endogenous peroxidation in vitro. Carbon tetra­chloride, however, reliably enhanced lipid peroxidation by its addition (l j x l / m l ) to microsomal incubations.Slater (197la) was able to show only slight stimulation of in vitro lipid peroxidation by trichlorofluoromethane.

The measurement of stimulation or depression of microsomal peroxidation was not easy as the degree of stimulation depended on the level of endogenous peroxidation. Duplicate analysis of at least three incubations was necessary to obtain any sort of reliable quantitative determination. Even so, the differences from the level of endogenous peroxidation were often too small to permit significance to be found. However, the changes found in most experiments with trichlorofluoro­methane were decreases in peroxidation, suggesting an anti-oxidant capacity. The mechanism is not likely to be a solvent effect, as this would be expected to cause increased peroxidation by exposing the microsomal phospholipids to metal ions and oxygen to an increasing extent. At the higher levels of addition of trichloro-

fTuoromethane to incubations (20 j x L / m L upwards) it is possible that the vaporised compound may have decreased the availability of oxygen to the microsomal suspensions by decreasing the partial pressure of oxygen. However, this would probably have played only a minor role. Thus,

from these results, it must be concluded that there is no evidence for the formation of a free radical analogous to that postulated to be formed from carbon tetrachloride. Therefore, if metabolism of trichloro- fluoromethane takes place it probably occurs by a different mechanism.

Indication that metabolism of the chlorofluoroalkanesmight occur, but at a level below the sensitivity of themethods discussed, was given by the fact that thesecompounds bound to cytochrome P~4f>0 (observed by thespectral interaction) in a manner identical to thechlorinated hydrocarbons, which are known to bemetabolically degraded. The affinities for cytochromeP-450 of the less volatile chlorofluoroalkanes arecomparable with the affinities measured for the chlorinatedhydrocarbons. If there is any basis for the correlationbetween the spectral constant (K ) and the Michaelis r sconstant (K ), metabolism of trichlorofluoromethane mmight be expected to occur at a rate only a little less than the rate of degradation of carbon tetrachloride.

Detection of in vitro metabolism of trichlorofluoro­methane would be made considerably easier if it were possible to separate metabolite and parent compound

before g.l.c. analysis. In addition, the use of P 6Cl] chlorofluoroalkanes in techniques similar to that described for studies of the metabolism of chlorinated ethanes and propanes (Van Dyke and Wineman, 1971) would permit detection of any dechlorination reaction that mayoccur with these chlorofluoroalkanes. Chlorofluoro-

l4 l4alkanes labelled with C would allow detection of C0oformed by oxidative degradation, as separation of carbondioxide and chlorofluoroalkane could be easily achievedby passing the expired air first through NaOH and thenthrough n-heptane.

The extent of metabolism of trichlorofluoromethane that might be expected, by analogy with the findings for carbon tetrachloride, is about 1%. Metabolism in excess of 0 . 2 % was not however detected. The reason for this may be found in the characteristics of the C-Cl bond in trichlorofluoromethane compared to the C-Cl bond in carbon tetrachloride. Table 1.3 shows that the bond energy term (calculated from the heat of formation) for a C-Cl bond in trichlorofluoromethane is 0.6% less than .in carbon tetrachloride, and that the C-Cl bond length is increased by about 3 % . This would suggest that trichlorofluoromethane would be dechlorinated as easily as carbon tetrachloride. However, the C-Cl bond dissociation energy in trichlorofluoromethane, calculated by Gregory (1 9 6 6) from the C-Cl dissociation energies measured for carbon tetrachloride and chlorotrifluoro­methane , is 9 • 5 % > greater than in carbon tetrachloride.

If t h i s extrapolated value is correct, the greater stability of the C-Cl bond in trichlorofluoromethane could account for the observed lack of metabolism.

The low toxicity of the chlorofluoroalkanes may thus be due to two factors, the lack of any significant metabolism and the very rapid clearance from the body, even after ingestion of the compounds orally. Although these compounds were detectable in adipose tissue after 24 hours, the amount present could only account for less than 0 .1% of the dose.

Toxicity.The low toxicity of trichlorofluoromethane and

1 ,2-dichloro-l,1 ,2 ,2-tetrafluoroethane was confirmed by experiments described in Chapter where no significant effect was found on body and tissue weights after a series of large, repeated doses.p In contrast to the massive accumulation of tri­glycerides in the liver during carbon tetrachloride intoxication (Ogata et al., 1968) and the increase inneutral lipid after halothane inhalation (Kunz et al.,1 9 6 6 ), trichlorofluoromethane (0.4 g/kg i.p. for 37 days) caused no accumulation of lipid, measured gravimetrically or colorimetrically (5•3 ) • However, a significantly lower level of * saturated* lipid was found in animals receiving trichlorofluoromethane. Ethyl oleate, used as the vehicle for trichlorofluoromethane, increased both the liver total-lipid and the liver * oleic acid1,

compared to rats receiving saline (containing sodium phenobarbital) , but did not affect the content of 1 saturated lipid*.

The decrease in * saturated lipid’ in the livers of rats that is attributable to trichlorofluoromethane may be interpreted in one of two ways: either the ratio of ’saturated* to *unsaturated1 lipid was altered, or the content of total lipid was decreased. Since the concentration of total lipid was not not significantly changed by trichlorofluoromethane, the.former case is more likely. This'is because, although the determination of total lipid is reliable and well defined, .’unsaturated* lipid, as measured, is an entity not capable of precise definition. However, a significant increase (P-C0.05) in total serum lipid has been reported (Hair, 1972) following the oral dosing of male rats with trichlorofluoromethane (lg/kg body weight/ day for 40 days). This increase was in part due to an increase in non-ester if ied fa:tty acids, which suggested the mobilisation of depot fat. Of relevance to these findings is the concentration of chlorofluoro­alkanes reported to occur in the adrenals (Davies, D.S., unpublished work), which may affect the release of adrenal hormones, resulting in an alteration in lipid mobilisation. The results reported here suggest that the degree of unsaturation of the liver lipid was raised by feeding trichlorofluoromethane and that this could be attributed to the use of ethyl oleate as vehicle. The significance of these several results is not clear, but

no deleterious effects of trichlorofluoromethane wereshown.

Binding with cytochrome P-450.As suggested above, the binding of chlorofluoro­

alkanes with cytochrome P-450 in microsomal preparations indicates that there is a potential for metabolism.The Kg value for trichlorofluoromethane is only a factorof two greater than the K for carbon tetrachloride.s

The breaks observed in the K plots were notsreproducible. This was probably due to the actual concentration of these volatile compounds in the microsomal suspension being lower than the calculated concentrations, due to small losses of vapour.However, when a break was found in the K plot, the two

S

intercepts with the abscissa gave K values approximately5

one order of magnitude apart, the higher value being the one obtained when no break was observed in the plot.The alteration in slope was not due to the presence of insufficient substrate, as even at the lowest substrate concentration used ( 1 0 0 nmol/ml), substrate exceeded cytochrome P-450 by a factor of 50. A possible interpretation of the two values obtained is that the substrate is binding at two sites of different affinities. Schenkman (1970) found this type of plot with the aniline-induced spectral change and suggested that the binding of the substrate to the cytochrome was facilitated by the presence of higher amounts of substrate due to a solvent effect.

The interaction of trichlorofluoromethane and other chlorofluoroalkanes and chlorinated hydrocarbons with aerobic cytochrome P-450 is predominantly with the phospholipid fraction, and the evidence for this is as follows. Removal of the phospholipid component by isooctane treatment abolished the type I spectrum without damaging the type II site, the aniline spectrum (type II) in fact being enhanced. Quite how much phospholipid was extracted by this procedure was not determined and attempts to reproduce the thin-layer chromatography results of Leibman and Estabrook (l97l) showed only one phospholipid spot, the'R^, value of which did not correspond to the R_ values given for either phosphatidylcholine or phosphatidylethanolamine.However, sufficient information to repeat exactly their system was not given in the paper. Their reported recovery of protein and cytochrome P-450 after isooctane extraction was poor, but by simply increasing the length and force of centrifugation it was possible to achieve recoveries of protein and cytochrome P-450 approaching 100%. Phospholipase C digestion of the microsomal preparation was also reported to prevent type I binding. The type I binding spectrum of trichloro- fluoromethane was modified in preparations digested with this enzyme, but was not completely abolished.

Further evidence for the lipid nature of the type I binding site is given by the spectral constants for various polyhalogenated compounds (Table 6 .l).Increasing substitution of chlorine for hydrogen in these

compounds (all type I substrates) resulted in an increasein the non-polar character and a decreased K value, i.e.s ’a higher affinity:

iriMch2ci2 1 0 . 3

CHC13 2.9CC14 1.3

CHC12-CH3 11.0CHC12~CH2C1 3.7CHC12-CHC12 2.8CHC12-CC13 .1.1CC13 -CC13 0.8

Similarly, carbon tetrachloride, hexachloroethane and octachloropropane showed increased affinity for the cytochrome with increased chlorine content (Kg values: l.»5i 0.8, 0.4 mM respectively). The effect of the substitution of fluorine for chlorine on the spectral constants was to cause an increase:

mMcci3-cci3 0.8CC12F-CC1F2 3-1c c i f2-c c i f2 8.1

Substitution of fluorine for hydrogen generally decreased the affinity for the cytochrome, one exception being CC1F -CC1F in comparison to CH Cl-CH Cl, as the

<L cL d* tL

latter gave no observable spectral interaction.Pretreatment of rats with phenobarbital, trichloro-

fluoromethane or carbon tetrachloride affected the K

values for trichlorofluoromethane and carbon tetrachloriCarbon tetrachloride pretreatment increased the K valuesfor both compounds. This may have been due to thepresence of low concentrations of carbon tetrachlorideretained in the microsomal preparations and inhibitinginteraction with other substrates. Pretreatment withtrichlorofluoromethane or phenobarbital, however,decreased the value of K for both carbon tetrachloridesand trichlorofluoromethane. It is possible that thisrepresents either the .formation of a slightly differentcytochrome V - k ^ O molecule or the modification -of theexisting cytochrome P-450.

As with many other parameters, species differenceswere found in K values (Table 6 .3)- However, all thesK values for trichlorofluoromethane lay within one s Jorder of magnitude, suggesting that the differences were probably of little significance. The only species of particular interest was the mouse, which showed the lowest spectral constant for trichlorofluoromethane.This fact, in addition to the high metabolic rates shown in mouse liver, suggests that the in vitro metabolism experiments with phenobarbital-induced mouse liver were the most likely to show positive evidence of metabolism, if it occurred. This reinforce the contention that reductive dechlorination of tri­chlorof luoromethane is probably not the mechanism of any possible metabolism in the systems used.

In terms of the type I nature of the spectral

interaction shown by trichlorofluoromethane with microsomal suspensions, it is interesting to note that this compound interferes with the type II spectrum of aniline, when added at a concentration of 2 mM. This is opposed to the findings of Schenkman (1970), who showed that the presence of low concentrations (2 . 5 kiM) of hexobarbital (type I) enhanced the spectral interaction produced by aniline Cl.9 and 20.5 mM). This interference by trichlorofluoromethane indicates that it shows some type II binding capacity, perhaps not spectrally observable, or that it modifies the type II site in some way.

Further evidence that trichlorofluoromethane andcarbon tetrachloride are not binding to the type I siteof cytochrome P-450 is provided by their formation ofreduced spectra closely similar to the spectrum of thecytochrome P-450-CO complex. Both these compoundsinterfered with the measurement of cytochrome P-450.The lowering of the cytochrome P-450-CO spectrum couldnot be reversed by prolonged bubbling with CO, suggestingthat the binding of carbon tetrachloride and trichloro-fluoromethane was of a similar affinity to the bindingof CO and probably occurred at the same site. Thisconflicts with the concept that the K value is asmeasure of affinity for cytochrome P-450, but Kutt et al.,(1 9 7 0) had previously described a similar lack ofcorrelation from results obtained with diphenylhydantoin.

—6This compound showed a very low ICg value (27 x 10 M)

but was found to be displaced easily by other substrates. The reverse would appear to be the case for trichloro- fluoromethane and carbon tetrachloride (Kg values approx.

_ o2 x 10 M) as binding of these compounds to Na^S^O^-reduced cytochrome P-450 was not reversed by CO, which

-6has a fK 1 of about 3 x 10 M (Omura and Sato, 1964a).S

In incubations of microsomal suspensions, cytochrome P-450 was degraded with time to a spectrally unobservable product. A substrate such as zoxazolamine afforded protection to the cytochrome, but did not affect the increased degradation due to trichlorofluoromethane or carbon tetrachloride. The latter compound caused a greater reduction in the concentration of cytochrome P-450 than did trichlorofluoromethane, and this difference was presumably due to peroxidative damage initiated by metabolism of carbon tetrachloride. Archakov et al. (1 9 7 2) have shown that substrates for demethylation block peroxidation and give binding spectra with cytochrome P-450. It seems possible therefore that the. inhibition of endogenous microsomal lipid peroxidation by chlorofluoroalkanes may be mediated in some way by their interactions with cytochrome P-450. This may explain the anomaly, reported in Chapter 4, concerning the apparent anti-oxidant effect of trichlorofluoro­methane .

Effects on drug-metabolising enzymes.In view of these spectral interactions in aerobic

and Na^S^O^-reduced microsomal suspensions, and the observed effect of halothane on drug-metabolising enzymes (Van Dyke and Rikans, 1970; Brown, 1971)? chlorofluoroalkanes might be expected to affect drug metabolism. However, the metabolism of type I binding substrates (biphenyl, ethylmorphine, hexobarbital and nitroanisole) was not affected. This may be explainedby the difference in the measurable affinities of

— 5 —4substrates (5 x 10 J to 5 x 10 M) and chlorofluoro-— 3 — 2alkanes (10 to 5 x 10 M). Competitive inhibition by

the chlorofluoroalkanes would have only a minimum effect on the enzyme activity. On the other hand, no solvent effect (similar to that of acetone), resulting in activation of enzyme activity (Anders, 1971)» was apparent.

The observed increase of hexobarbital sleeping times cannot be explained therefore by inhibition of metabolism. It seems likely that.this effect may be centrally mediated. Some weak local anaesthetic activity was noted in experiments involving intra- peritoneal injection of chlorofluoroalkanes, which resulted in temporary paralysis of the animal1s hind limbs.

Metabolism of some type II substrates in vitro was altered by addition of chlorofluoroalkanes to the incubation media. Thiabendazole 5-hydroxylation was unaffected. Zoxazolamine hydroxylation was apparently significantly inhibited, but this may possibly have been

due to the decreased availability, or increased destruction, of cytochrome P-450, due to binding of trichlorofluoromethane or carbon tetrachloride, over the long period of the incubation (40 to 60 minutes).Unfortunately, it was not possible to reduce the incubation time to less than 40 minutes, as the rate of hydroxylation was relatively low. Aniline and acetanilide 4-hydroxylations were, however, increased by the addition of trichlorofluoromethane. Van Dyke and Rikans (1970) observed a similar effect on aniline hydroxylation, using either halothane or 1 ,1 ,1-trichloro- ethane. The activation of aniline and acetanilide metabolism by trichlorofluoromethane did not appear to be a solvent effect, as addition of ethanol caused a 30% inhibition of this activity, and 1 ,1 ,2-trichloro-1 ,2 ,2 -trifluoroethane did not produce such large increases. Anders (1 9 6 8) found enhancement of aniline 4-hydroxylation by acetone (0.045 - 1.8 M), but not by similar ketones, and suggested that the mechanism of activation was some alteration of the enzyme, and not simply a solvent effect.

Spectral interactions as an indication of metabolism.The type of reduced binding spectrum produced with

trichlorofluoromethane and carbon tetrachloride has been found with a number of compounds, including hydralyzines such as *Nardilf (Netter, 1972), fluorene and 2-nitro- propane (Ullrich, 1973)i ethyl isocyanide and carbon monoxide (Omura and Sato, 1964a). Aniline gives an

extremely weak interaction, of this type (imai and Sato,1967).

In the case of trichlorofluoromethane and carbon tetrachloride, the formation of the reduced spectrum was time-dependent, and slow relative to CO binding (Omura et al., 1 9 6 5 )1 possibly suggesting an electron-transferreaction or an allosteric alteration of the cytochrome P-450. The latter alternative would be the probable mechanism if this spectral interaction was found to occur using crystalline haem. The formation of the reduced spectrum was not abolished by any of the treatments to remove phospholipid from the microsomal suspension; phospholipase C reduced the rate of formation of the peak, which suggests that the digestion of the microsomal phospholipid made access to the haemoprotein more difficult, whereas isooctane may increase the accessibility by more complete removal of the phospholipid. These observations indicate that trichlorofluoromethane binds to a second site within cytochrome P-450 similar to the type II binding site observed in aerobic preparations. The activation of aniline and acetanilide 4-hydroxylations would indicate that this second site is close to, but not identical to, the type II substrate binding site, as otherwise inhibition of all type II drug metabolising activity would be expected.Another possibility is that trichlorofluoromethane binds to a site which allosterically affects the type II binding site of cytochrome P-450.

The meaning and importance of the reduced binding spectra observed with trichlorofluoromethane and 1 ,1 ,2-

trichloro-1,2,2-trifluoroethane is open to considerable speculation. Similar spectral changes have been observed in microsomal preparations incubated with substrates and NADPH. Piperonyl butoxide (Philpot and Hodgson, 197l) 5 safrole and isosafrole (Lake and Parke, 1972) and hexobarbital (Estabrook et al., 1971) incubated with microsomal preparations in vitro, with an NADPH generating system, resulted in the formation of spectral intermediates showing maxima between 440 nm and 4.55 11m • These spectra are believed to be the result ofthe binding of metabolites to cytochrome P-450.Schenkman et al. (1972) have shown that active metabolismof SKF 525A in vivo and in vitro resulted in the formation of a stable oxygenated complex of cytochrome P-450, with an absorption maximum at 455 nm. This -complex may be responsible for the known non-competitive inhibition of drug metabolism by SKF 525A.

The implication of this work and recent observations on the reduced binding of carbon tetrachloride is that trichlorofluoromethane may in fact be metabolised to a small extent. Reiner and Uehleke (unpublished work) have shown that the anaerobic formation of the NADPH- reduced spectrum of carbon tetrachloride is almost exactly paralleled by the formation of chloroform, and they have postulated a two-step mechanism, involving the anaerobic binding of carbon terachloride to the cytochrome and the first electron transfer, followed by metabolism by the mixed-function oxygenase and the second

electron transfer, chloroform being released at this step. The trichlorofluoroinethane-cytochrome P-450 reduced spectral intermediate may thus indicate the possibility of the formation of dichlorofluoromethane under anaerobic and reducing conditions.

The formation of the reduced spectrum of tri­chlorof luoromethane in microsomal preparations, on the other hand, may reflect a ligand interaction as the first step in metabolism, which cannot be followed by the formation of free metabolite in the case oftrichlorofluoromethane. The * spectral K 1 (Figure 6.15)mof cytochrome P-450 for trichlorofluoromethane may thus be an expression of the affinity of cytochrome P-450 for this compound in reducing conditions or of the concentration of trichlorofluoromethane methane required for half-maximal rate of metabolism.

Ligand interactions producing similar spectral changes have been observed between ferrihaemoglobin and the chlorinated methanes (Bucher and Brown, 1971)- Various polyhalogenated methanes, including dichloro- difluoromethane, have been combined with reduced vitamin B (cobalamin)(Wood et al., 1968) to form crystallisablederivatives. These complexes were shown to involve a ligand interaction between the halogenated methanes and the cobalt atom of cobalamin. Photolysis of the chlorinated methane derivatives under hydrogen yielded chloromethane and hydrogen chloride. Photolysis of the chlorofluoroalkane derivative of cobalamin was not

reported.Thus it might be postulated that metabolism of

trichlorofluoromethane may occur by two possible routes

J CC12F] .s - a

CHCl^F✓CCl^F

Any metabolism of trichlorofluoromethane in aerobic conditions, such as those described in Chapter 4, may result in some oxidised product (fX !), and not dichloro fluoromethane. It may only, be possible to form this latter metabolite under completely anaerobic conditions

'From these studies,.the low toxicity of the chlorofluoroalkanes would appear to be due to a a combination of the rapid clearance of the compounds from the tissues and the very low level, or lack, of metabolism.

However, the binding of the chlorofluoroalkanes to cytochrome P-450, and the effects on in vitro drug metabolism, lead one to the suggestion that further studies should be undertaken to determine whether these effects could occur in vivo.

Further metabolic studies require, ideally, the use of radioactively-labelled compounds. It would seem to be of particular interest to determine whether any metabolism occurs in anaerobic, reducing conditions.If metabolism does not occur under these conditions, trichlorofluoromethane may prove to be a useful tool in separating the two steps of Uehleke!s postulated mechanism for the metabolism of carbon tetrachloride.

REFERENCES

References.

Allott, P.R., Steward, A. and Mapleson, ¥.¥. (l97l)- Brit. J. Anaesth., 43, 913~9l8.

Alvares, A.P., Schilling, G., Levin, ¥. and Kuntzman, R(1971)• J. Pharmacol, exp. Ther., 1 7 6, 1-10.

Anders, M.¥. (1 9 6 8). Arch. Biochem. Biophys., 126, 269“273.

Anders, M.¥. (l97l)- Ann. Rev. Pharmacol., 1 1 , 37-56.Archakov, A.I., Karuzina, I.I., Bokhon’ko, A.I.,

Alexandrova, .T.A. and Panchenco, L.F. (1972). Biochem. Pharmacol., 21, 1595-l602.

Barker, E.A., Arcasoy, M. and Smuckler, E.A. (1969)- Agents and Actions, _1 , 27-34.

Barnett, J.E.G., Jarvis, ¥.T.S. and Munday, K.A. (1 9 6 7) Biochem. J ., 103, 699-704.

Barnsley, E.A. (1 9 6 6). Biochem. J ., 100, 362-372.Barton, J.D.M. (1959)- Lancet, jl, 1097-Baselt, R.C. and Cravey, R.H. (1 9 6 8)..J. Forensic Sci.,

13, 407-410.Bass, M. (1970). J. Am. Med. Assn., 212, 2075-2079- Belfrage, S., Ahlgren, I. and Axelson, S. (1966).

Lancet, ii, l466-l467-Berinan, M.L. and Bochantin, J.F. (1970). Anesthesiology

3 2 , 5 0 0-5 0 6 .Blake, D.A. and Cascorbi, H.F. (1970). Anesthesiology,

3 2 , 5 6 0 .Bloom, R.J. and ¥esterfeld, ¥.¥. (l97l)- Arch. Biochem.

Biophys., 145, 669-675-

Boethner, E.A. and Muranko, H.J. (1969)- Am. Ind. Hyg. Assn. J ., 30, 437-442.

Booth, H.S. and Bixby, E.M. (1932). Ind. Eng. Chem., 24, 637-641.

Booth, J., Boyland, E. and Sims, P. (l96l). Biochem. J.,72, 5 1 6-5 2 4 .

Bourne, P.G. and Murphy, W.R. (1 9 6 9)* J. Soc. Cosmetic Chemists, 20, 525-537-

Boyland, E. and Chasseaud, L.F. (1 9 6 9). Biochem. J .,115, 985-991.

Br a che t - Li erma in, A., Ferrus , L. and CarofT, J. (l97l).«J. Chromatog. Sci., 2i 49-53-

Bray, H.G. and James, S.P. (1958). Biochem. J., 6 9 , 24P.Bray, H.G., Thorpe, W.V. and Vallance, D.K. (1952).

Biochem. J ., 51, 193-201.Brenner, C. (1937 ) - J. Pharmacol., 59, 176-l8l.Brodie, B.B. and Axelrod. J. (1948). J. Pharmacol, exp.

Ther., 94, 22-28.Brody, G.L. and Sweet, R.B. (1963)- Anesthesiology, 24,

29-37-Brook, R.J. and Joyner, B.D. (1 9 6 6 ). J. Soc. Cosmetic

Chemists, 17, 401-4l4.Brown, B.R. (l97l)- Anesthesiology, 35, 241-246.Brown, B.R. and Vandam, L.D. (l97l)- Ann. N.Y. Acad. Sci.

179, 235-243.Bryce, H.G. (1964) in Fluorine Chemistry, (Simons, J.H.,

ed.), vol. V, pp. 295-498, Academic Press, N.Y. and London.

Bucher, D.J. and Brown, W.D. (l97l)« Biochemistry, 10,4239-4246.

Burn, J.H., Epstein, H.G. and Goodford, P.J. (1959).Brit. J. Anaesth., 31, 518-529.

Burnap, T.K. , Galla, S.J. and Vandam, L.D. (1958). Anesthesiology, 1 9 , 307-320.

Burns, J. and Snow, G.A. (1 9 6 1). Brit. J. Anaesth., 33, 102-103.

Butler, R.A. and Hill, D.W. (1 9 6 1). Nature (London),1 8 9 , 488-4 8 9 .

Butler, T.C. (l96l). J. Pharmacol, exp. Ther., 134,311-319.

Carpenter, C.P., Smyth, H.F. and Pozzani, U.C. (1949).J. Ind. Hyg. Toxicol., 31, 343-346.

Cascorbi, H.F. and Blake, D.A. (l97l)« Anesthesiology,35, 493-495.

Cascorbi, H.F., Vesell, E.S., Blake, D.A. and Helrich, M (1971a). Clin. Pharmacol. Ther., 12, 50-55-

Cascorbi, H.F., Vesell, E.S., Blake, D.A. and Helrich, M (1971b). Ann. N.Y. Acad. Sci., 179, 244-248.

Cessi, C., Colombini, C. and Mameli, L. (1 9 6 6). Biochem. J ., 101, 46-47C.

Chaplin, M.D. and Mannering, G.J. (1970). Mol. Pharmacol 6 _ , 6 31-640. ■ •

Chenoweth, M.B. and Hake, C.L. (1 9 6 2). Pharmacol. Rev.,2 , 363-398.

Chenoweth, M.B., Van Dyke, R.A. and Erley, D.S. (1962). Nature (London), 194, 575-576.

Chichugova, T.N., Rabovskii, G.V. and Zalesskii, V.N.(1964). Gazov Khromatog. , jL, 132-134.

Clayton, J.W. (1 9 6 2). J. Occup. Med., 4, 262-273*

Clayton, J.W. (1966) in Handbuch der exp. Pharmakol. (Eichler, 0., Farrah, A., Herken, H. and Welch,A.D., eds.), vol. 20, pp. 459-500, Springer, Berlin.

Clayton, J.W. (1967a). Fluorine Chem. Rev., jL, 197-252.Clayton, J.W. (1 9 6 7b). J. Soc.. Cosmetic Chemists, 1 8 ,

333-350. . .Clayton, J.W. (1 9 6 8 ) in Toxicity of Anesthetics (Fink,

B., ed.), Williams and Wilkins, Baltimore.Clayton, J.W., Delaphane, M.A. and Hood, D.B. (i9 6 0 ).

Am. Ind. Hyg. Assn. J., 21, 3 8 2-3 8 8 .Clayton, J.W., Hood, D.B., Nick, M.S. and Waritz", R.S.

(1966), Am. Ind. Hyg. Assn. J ., 27, 234-238.Clemons, C.A. and Altshuller, A.P. (1 9 6 6). Anal. Chem.,

38, 133-136.Cohen, E.N. (1969)- Anesthesiology, 31, 5 6 0-5 6 5 .Cohen, E.N. (l97l)« Anesthesiology, 35, 193-202.Cohen, E.N. and Brewer, H.W. (1964). J. Gas Chromatog.,

1 2 6 l-2 6 2 .Cohen, E.N. and Hood, N. (-1969)- Anesthesiology, 31,

553-559.Cohen, E.N., Bellville, J.W., Budzikiewicz, H. and

Williams, D.H. (1 9 6 3). Science, l41, 899-Cohen, E.N., Brewer, H.W., Bellville, J.W. and Sher, R.

(1 9 6 5 ). Anesthesiology, 26, l40-153*Conney, A.H., Trousof, N. and Burns, J.J. (i9 6 0 ). J.

Pharmacol, exp. Ther., 128, 333-339*Cottrell, T.L. (1958). The Strengths of Chemical Bonds,

2nd edn., Butterworths, London.Creaven, P.J., Parke, D.V. and Williams, R.T. (1 9 6 5).

Biochem. J., 9 6 , 8 7 9-8 8 7 .Curry, A.S., Hurst, G., Kent, N.R., and Powell, H. (1 9 6 2).

Nature (London), 1951 603-604.

Daly, J.W., Guroff, G., Udenfriend, S. and Witkop, B.(1968). Biochem. Pharmacol., 17, 31-36.

Daniel, J.W. (1 9 6 3)- Biochem. Pharmacol., 12, 795-802.Davis, D.C., Schroeder, D.H., Gram, T.E., Reagen, R.L.

and Gillette, J.R. (l97l). J.Pharmacol, exp. Ther.,177, 5 5 6-5 6 6 .

De Becker, J., Hupin, C. and Massart, N. (1 9 6 7). J. Pharm. Belg., 22, 188-198.

Degkwitz, E., Ullrich, V., Staudinger, H. and Rummel, W.(1969)« Hoppe-Seyler1s Z. Physiol. Chem., 350, 547-553-

Dingell, J.V.. and Heimberg, M. (1968). Biochem.Pharmacol. , 17, 1269-1278'.

Dodge, J.T. and Phillips, G.B. (1 9 6 6). J . Lipid Res. , 7.,387-395.

Dollery, C.T., Draffan, G.H., Davies, D.S., Williams,P.M. and Conolly, M.E. (1970). Lancet, ii, Il64-ll66.

Downer, H.D., Galloway, R.W., Horwich, L. and Parke, D.V. (1.970) J. Pharm. Pharmac., 22, 479-487-

Downing, R.C. and Madinabeitia, D. (i9 6 0 ). Aerosol Age,5., 25-27.

Duve, C. de, Pressman, B.C., Gianetto, R., Wattiaux, R. and Appelmans, F. (1955). Biochem. J., 60, 6o4-6l7-

Ehrner-Aamuel, H. (1 9 6 3)- Proc. l4th Int. Congr. Occupational Hlth., 62, 394-397-

Eling, T.E. and DiAugustine, R.P. (l97l)« Biochem. J .,123, 539-549.

Ellis, M. (l96l). Brit. Med. J., ii, 250-252.Epstein, S.S., Andrea, J., Clapp, P., Mackintosh, D. and

Mantel, N. (1967a). Toxicol. Appl. Pharmacol., 11,442-448.

Epstein, S.S., Joshi, S., Andrea, J., Clapp, P., Falk,H. and Mantel, N. (1 9 6 7b). Nature (London), 214, 5 2 6 -5 2 8 .

Esposito, G.G. and Swann, M.H. (1 9 6 7). J. Paint Technol. 29, 338-340.

Estabrook, R.W. , Cooper, D. and Rosenthal, 0. (1963)-Biochem. Z., 338, 741-755-

Estabrook, R.W., Hildebrandt, A.G., Baron, J., Netter,K.J. and Leibman, K. (l97l)- Biochem. Biophys. Res. Comm., 42, 132-139-

Feuer, G. and Granda, V. (1970). Toxicol. Appl. Pharmacol.1 6 , 626-637.

Florini, J.R. and Buyske, D.A. (1959). Arch. Biochem. Biophys., 79, 8-12.

Folch, J., Lees, M. and Sloane-Stanley, G.H. (1957).J. Biol. Chem., 226, 497-509-

Foris, A. and Lehman, J.G. (1 9 6 9). Sep. Sci., _4i 225-242.Fowler, J.S.L. (1969a). Brit. J. Pharmacol., 35, 530-542.Fowler, J.S.L. (1 9 6 9b). Brit-. J. Pharmacol., 37, 733-737-Franz, J. and Cole, B.T. (1 9 6 2). Arch. Biochem. Biophys.

96, 3 8 2-3 8 5 .Frings, C.S. and Dunn, R.T. (1970). Am. J. Clin. Path.,

53, 89-91.

Gadsen, R.H. and McCord, W.M. (1964). J. Gas Chromatog. 2 , 7-11.

Gage, J.C. (1 9 6 3). Brit. J. industr. Med., 20, 248-249- Ghoshal, A.K. and Recknagel, R.O. (1 9 6 5). Life Sci., _4*

1521-1530.

Gilbert, D . and Goldberg, L. (1965)- Food Cosmet.Toxicol., 2? 417-432.

Gillette, J.R. (1971a). Metabolism, 20, 215-227.Gillette, J.R. (1971b). Ann. N.Y. Acad. Sci., 179, 43-66.Gillette, J.R., Davis, D.C. and Sasame, H.A. (1972).

Ann. Rev. Pharmacol., 12, 57-84.Glende, E.A. (1972). Biochem. Pharmacol., 21 , 2131-2138.Glockler, G. (1959). J . Phys. Chem., 63, 8 2 8-8 3 2 .Goldberg, M.A. and Netzbandt.. (19 6 6 ) • U.S.Patent 3,288,681.Goldman, P. (1965). J. Biol. Chem., 240, 3434-3438.Goldman, P. (1 9 6 9). Science , 1.64, 1123-1130.Goldman, P., Milne, G.W.A._and Pignataro, M.T. (1 9 6 7)-

Arch. Biochem. Biophys., 11 8 , 178-184.Gordis, E. ( 1 9 6 9)• J. Clin. Invest., 48, 203-209.Gottlieb, A.A., Fujita, Y., Udenfriend, S. and Witkop, B.

(1 9 6 5). Biochemistry, Jt* 2507-2513-Greenberg, L.A. and Lester, D. (1950). Arch. Ind. Hyg.

Occup. Med., 2? 345-347.Gregory, N.L. (1 9 6 6). Nature (London), 212, l460-l46l.Grenby, T.H. and Young, L. (i9 6 0 ). Biochem. J ., 7 5 , 28-33-Guroff, G., Kondo, K. and Daly, J. (1 9 6 6). Biochem.

Biophys. Res. Comm., 25, 622-628.Gvosdovich, T.N. and Jashin, J.I.’ (1970)- J. Chromatog.,

49, 36-39-

Hair, C. (1972). B.Sc. Project, University of Surrey.Hanes, C.S. (1932). Biochem. J ., 26, l406-l421.Hansen, V. (1970). Acta anaesth. Scand., l4, 1-3- Harris, M.C. (1971)- Annals of Allergy, 29, 250-256.

Hartmann, K-U., and Heidelberger, C. (1 9 6 1). J. Biol. Chem., 236, 3006-3013.

Hathway, D.E. (1970) in Foreign Compound Metabolism in Mammals, vol.I, Chemical Society, London.

Heni, N. (l97l)- Experientia, 27, 777-778.Henne, A.L. (1937). J . Am. Chem. Soc. , 5_9, l400-l401.Heppel, L.A. and Porterfield, V.T. (1948). J. Biol. Chem.

1 7 6 , 763-769.Hine, J. (1950)- J. Am. Chem. Soc., 52, 2438-2445-Hodge, H.C. and Smith, F.A. (1 9 6 5) in Fluorine Chemistry,

(Simons, J.H., ed.), vol. IV, Academic Press, N.Y. and London.

Hodge, H.C., Smith, F.A. and Chen, P.S. (19 6 3 ) inFluorine Chemistry, (Simons, J.H., ed.), vol. Ill, Academic Press, N.Y. and London.

Holaday, D.A., Rudofsky, S. and Treuhaft, P.S. (1970). Anesthesiology, 33, 579-593-

•Holtzman, J.L., Gram, T.E., Gigon, P.L. and Gillette, J.R (1968). Biochem. J., 110, 407-4l2.

Hughes, G.M.K. and Saunders, B.C. (1954). J. Chem. Soc. (London), 1954, 4630-4634.

Hutson, D.H. (1970) in Foreign Compound Metabolism in Mammals, vol. I, Chemical Society, London.

Imai, Y. and Sato, R. (1967)- J. Biochem. (Tokyo), 6 2 , 239-249.

Irish, D.D. (1 9 6 2) in Industrial Hygiene and Toxicology (Patty, F.A., ed.), Interscience, N.Y. and London.

Jack, D. (1971). Brit. Med. J ., ii, 708-709.Jenkins, L.J., Jones, R.A., Coon, R.A. and Siegel, J.

(1970). Toxicol Appl. Pharmacol., 1 6 , 133-142.Johnston, C.I. and Mendelsohn, F. (1971). Aust. N.Z. J.

Med., 2 , 171-173.Jondorf, W.R. , Parke, D.V. and Williams, R.T. (1957)-

Biochem. J. , 6 _5, 14-15P.Juchau, M.R., Cram, R.L., Plaa, G.L. and Fouts, J.R.

(1 9 6 5 ). Biochem. Pharmacol., 14, 473-482.

Kamataki, T., Kitada, M. and Kitagawa, H. (l97l). Chem. Pharm. Bull.', 19, 1749-1750.

Kaufman, S. (1 9 6 1). Biochim. Biophys. Acta, 51, 619-621.Keown, IC.K. and Bingham, H.G. (1 9 6 9 ). Anesth. Analg.,

48, 710-714.Kety, S. (i9 6 0 ). Meth. med. Res., 228-236.Klein, N.C. and Jeffries, G.H. (19663 - J. Am. Med. Assn.,

197, 1037-1039.Konig, H. (I.9 6 7 )- Fres. Z. Anal. Chem., 232,. 427-432.Krisch, K. and Staudinger, H. (1 9 6 1). Biochem. Z., 334,

312-327.Krisch, K., Staudinger, H. and Ullrich, V. (1964). Life

Sci., 3, 97-102.Kubler, H. (1963)- J. Soc. Cosmetic Chemists, l4, 341-351-Kuntzman, R. (196 9)- Ann. Rev. Pharmacol., _9 > 21-36.Kunz, W.,Schaude,| G. , Schimassek, H. , Schmid, W. and

Siess, M. (1 9 6 6). Proc. Eur. Soc. Study of Drug Toxicity, _7i 138-153-

Kutt, H., Waters, L. and Fouts, J.R. (1970). Chem.-Biol. Interactions, 2, 195-202.

Lake, B.G. and Parke, D.V. (1972). Biochem. J ., 127,23P.

Lai, II., Puri, S.K. and Fuller, G.C. (1970). Toxicol.Appl. Pharmacol., l6 , 35-39-

Larsen, E.R. (1963)- (to the Dow Chemical Company)U.S. Pat. 3,104,202.

Larsen, E.R. (1969)- Fluorine Chem. Rev. , 3.1 1-44.Lebeau, P. and Damiens, A. (1926). Compt. Rend., 1 8 2 ,

1340-1342.Leibman, K.C. and Estabrook, R.W. (l97l)- Mol. Pharmacol.

7_, 2 6-3 2 .Lester, D. and Greehberg, L.A. (1950). Arch. Ind. Hyg.

Occup. Med., 2i 335-344.Liebecq, C. and Peters, R.A. (1949). Biochim. Biophys.

Acta., 3, 215-230.Linde, H.W. and Bruce, D.L. (.1 9 6 9). Anesth. , 30, 3 6 3-3 6 8 .Lindebaum, J. and Leifer, E. (1 9 6 3 ). New Eng. J. Med.,

2 6 8 , 5 2 5-5 3 0 .Lineweaver, H. and Burk, D. (1934). J. Am. Chem. Soc.,

5 6 , 6 5 8-6 6 6 .Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J.

(l95l). J. Biol. Chem., 193 * 265-275*Lu, G., Ling, J.S.L. and Krantz, J.C. (1953). Anesthesiology,

14, 466-472.Lucas, G.H.W. (1928). J. Pharmacol, exp. Ther., 34,

223-237.

May, H.E. and McCay, P.B. (1 9 6 8). J . Biol. Chem., 243, 2296-2305.

Mazze, R.I., Shue, G.L. and Jackson, S.H. (l97l)«J. Am. Med. Assn., 216, 278-288.

Michaelson, J.B. and Huntsman, D.J. (1964). J . Med. Chem., 7, 378-379.

Midgley, T. and Henne, A.L. (1930). Ind. Eng. Chem.,22 , 542-545.

Mirosevic-Sorgo, P. and Saunders, B.C. (1959). Tetrahedron,5, 38-43.

McCollister, D.D., Beamer, W.H., Atchison, G.J. and Spencer, M.C. (l95l). J. Pharmacol, exp. Ther., 1 0 2 , 112-124.

McLean, A.E.M. and McLean, E.K. (1965)- Biochem. J .,97, 31P.

McLean, A.E.M. and McLean, E.K. (1966). Biochem. J .,1 0 0 , 564-571.

Nakanishi, S., Masamura, E., Tsukada, M. and Matsumura, R.(l97l). Jap. J. Pharmacol., 21, 303-309-

National Halothane Study (1 9 6 6). J. Am. Med. Assn., 197, 775-788.

Netter, K.J. (1972), in the press.Netter, K.J. and Seidel, G. (1964). J. Pharmacol, exp.

Ther., l46, 6 1-6 5 .

Nikitenko, T.K. and Tolgskaya, M.S. (1965)- Gigiena Truda i Prof. Zabolevaniya, 2? 37-44.

Nowill, W.K. (1970). Anesth. Analg., 49, 355-360.Nuckolls, A.H. (1933). Underwriters Lab. Rept. M.H. 2375*.

Ogato, M., Tomokuni, K. and Watanabe, S. (1968). Ind. Health, 6 116-119-

Omura, T. and Sato, R. (1964a). J. Biol. Chem., 239, 2370-2378.

Omura, T. and Sato, R. (1964b). J . Biol. Chem., 239, 2379-2385.

Omura, T. , Sato, R., Cooper, D.Y., Rosenthal, 0. andEstabrook, R.W. (1965)- Fed. Proc., 24, Il8l-ll89»

Panner, B.J., Freeman, R.B., Roth-Moyo, L.A. andMarkowitch, W. (1970). J. Am. Med.Assn., 2l4,86- 90.

Parke, D.V. (1968). The Biochemistry of Foreign Compounds, Pergamon Press Ltd., Oxford.

Paterson, J.W., Sudlow, M.F. and Walker, S.R. (l97l)« Lancet, ii, 565-568.

Paul, B.B. and Rubinstein, D. (1963)- J. Pharmacol, exp. Ther., l41, l4l-l48.

Paulet, G., Chevrier, R., Paulet, J., Duchene, M. and Chappet, J. (1969)« Arch. Mai. Prof. Med. Trav.Secur. Soc., 30, 101-120.

Pauling, L. (i960). The Nature of the Chemical Bond, 3*'d edn., Cornell Univ. Press, Ithaca, N.Y.

Petrova, M.P., Alekseeva, D.D. and Meshcheryakova, A.N. (1968). Zh. analit. Khim., 23, 1101-1103.

Philpot, R.M. and Hodgson, E. (l97l)- Life Sci., 10, 503-512.

Pickthall, J. (1964). Pharm. J., 193, 391-395-Prendergast, J.A., Jones, R.A., Jenkins, L.J. and

Siegel, J. (1967)- Toxicol. Appl. Pharmacol., 10, 270-289.

Purchase, I.F.H. (1963)- Nature (London), 198, 8 9 5 -8 9 6 .

Quevauviller, A. (1 9 6 5)- Prod. Probl. Pharm., 20, 14-29-Quevauviller, A., Chaigneau, M. and Schrenzel, M. (1963)-

Ann. Pharm. Franc., 21, 727-734.Quevauviller, A., Schrenzel, M. and Huyen, V.N. (1964).

Therapie, 191 247-263•

Ratcliffe, D.B. and Targett, B.H. (1969)- Analyst, 94, 1028-1033.

Raventos, J. (1956). Brit. J. Pharmacol., 11, 394-409-jRecknagel, 0. (1967). Pharmacol. Rev., 19? 145-208.Reed, W.H. (1 9 6 1). U.S. Patent 2 ,9 6 8 ,6 2 8 .Rehder, K., Forbes, J., Alter, H., Hessler, 0., and

Stier, A. (1 9 6 7)- Anesthesiology, 28, 711-715-Reiner, 0. and Uehleke, H. (l97l)- Hoppe-Seyler1s Z.

Physiol. Chem., 352, 1048-1052.

Reinhardt, C.F., Azar, A., Maxfield, M.E., Smith, P.E. and Mullin, L.S. (1971a). Arch. Environ. Health,2 2 , 265-279.

Reinhardt, C.F., McLaughlin, M., Maxfield, M.E.,Mullin, L.S. and Smith, P.E. (l971b)- Am. Ind.Hyg. Assn.' J., 32, 143-152.

Remmer, H., Schenkman, J.B., Estabrook, R.W., Sasame, H., Gillette,J . Cooper, D.Y., Narasimhulu, S. and Rosenthal, 0. (1 9 6 6). Mol. Pharmacol., _2» 187-190.

Remmer, H., Estabrook, R.W., Schenkman, J. and Greim, H.(1968). Naunyn Schmeideberg1s Arch. Pharm. exp. Path 259, 9 8-1 1 6 .

Reynolds E.S. (1963). Fed. Proc., 22, 1253 Abstr.Robbins, B.H. (1946). J. Pharmacol, exp. Ther., 8 6 ,

197-204.

Sasame, H.A., Castro, J.A. and Gillette, J.R. (1968). Biochem. Pharmacol., 17, 1759-1768.

Saslaw, L.D., Corwin, L.M. and Waravdekar, V.S. (1 9 6 6 ). Arch. Biochem. Biophys.,, 114, 6 l-6 6 .

Saunders, B.C. and Stacey, G.J. (1948). J. Chem. Soc. (London), 1948, 1773-1779.

Saunders, B.C., Stacey, G.J. and Wilding, I.G.E. (1949). J. Chem. Soc. (London), 1949 , 773-777.

Sawyer, D.C., Eger, E.I., Bahlman, S.H. and Cullen, B.F. (1970). Fed. Proc., 29, 354 Abstr.-

Sayers, R.R., Yant, W.P., Chornyak, J. and Shoaf, H.W. (1930). U.S. Bur. Mines Rept. Invest., 3013 -

Scheig, R. and Klatskin, G. (1 9 6 9). Life Sci., 2? 855-865.

Schenkman, J.B. (1970). Biochemistry, J9, 2081-2091-Schenkman, J.B. and Sato, R. (1 9 6 8). Mol. Pharmacol., 4

613-620.Schenkman, J.B., Remmer, II. and Estabrook, R.W. (1967)-

Mol. Pharmacol., 3., 113-123-Schenkman, J.B., Wilson, B.J. and Cinti, D.L. (1972).

Biochem. Pharmacol., 21 , 2373-2383-Scholler, K.L. (1970). Brit. J. Anaesth., 42, 6 0 3-6 0 5 .Scholz, J. (1962). Berlin Aerosol-Kongress, 4_, 420-429-Seawright, A.A. and McLean, A.E.M. (1967)- Biochem. J .,

105, 1 0 3 5-1 0 6 0.Silverglade, A. (l97l)- J. Am. Med. Assn., 215i ll8 .Silverman, P., Marcis, J. and Schmidt, C.N. (1 9 6 6).

Chem. Spec. Mf r . Assn., Proc. Ann. Meet., 531 31-39Sims, P. and Grover, P.L. (1965)- Biochem. J ., 951

156-160..Slater, T.F. (1965)- Biochem. Pharmacol., l4, 178-l8l. Slater, T.F. (1 9 6 6). Nature (London), 209, 36-40.Slater, T.F. (1972). Free Radical Mechanisms in Tissue

Injury, Pion Ltd., London.Slater, T.F. and Sawyer, B.C. (1971a). Biochem. J . , 123 ?

8 0 5 -8 1 4 .Slater, T.F. and Sawyer, B.C. (1971b). Biochem. J., 123

8 1 5-8 2 1 .Slater, T.F. and Sawyer, B.C. (1971c). Biochem. J., 123

8 2 3-8 2 8 .Sokoloff, L. (1957) in New Research Techniques of

Neuroanatomy (Windle, W., ed.), C.C. Thomas, Springfield.

Sperry, W.M. and Brand, F.C. (1955)- J - Biol. Chem., 213 69-76.

Stephan, C.R., Margolis, G., Fabian, L.W. and Bougeois- Gavardin, M. (1958)* Anesthesiology, 19, 770-781.

Stewart, R.D., Swank, J.D., Roberts, C.B. and Dodd, H.C. (1963)- Nature (London), 1 9 8 , 696-697-

Stewart, R.D., Dodd, H.C., Gay, H.H. and Erley, D.S.(1970a). Arch. Environ. Health, 20, 64-71-

Stewart, R.D., Baretta, E.D., Dodd, H.C. and Torkelson, T.R. (1970b). Arch. Environ. Health, 20, 224-229-

Stier, A. (1964a). Naturwissenschaften, 5_1, 65-Stier, A. (1964b). Biochem. Pharmacol., 131 1544.Stier, A. (1968). Anesthesiology, 29, 3 8 8-3 8 9 *Stier, A., Alter, H., Hessler, 0. and Rehder, K. (1964).

Anesth. Analg., 43, 723-728.

Taves, D.R., Fry, B.W., Freeman, R.B. and Gillies, A.J.(1 9 7O). J. Am. Med. Assn., 2l4, 91-95-

Taylor, G.J. and Harris, W.S. (1970). J. Am. Med. Assn.,214, 8 1-8 5 .

Thomson, A.E.R., Barnsley, E.A. and Young, L. (1963)- Biochem. J ., 8 6 , 145-152.

Tygstrup, N. (1963)- Lancet, ii, 466.

Ullrich, V. (1973)- J. Drug. Metab., in the press.

Vandam, L.D. (1963) . Anesthesiology, 24, 109-110.Van Dyke, R.A. (1 9 6 6). J. Pharmacol, exp. Ther., 154,

364-369.Van Dyke, R.A. and Chenoweth, M.B. (1 9 6 5a). Anesthesiology

2 6 , 348-357.Van Dyke, R.A. and Chenoweth, M.B. (19 6 5b). Biochem.

Pharmacol., l4, 6 0 3-6 0 9 .Van Dyke, R.A. and Rikans, L.E. (1970). Biochem.

Pharmacol., 19 , 1501-1502.Van Dyke, R.A. and Wineman, C.G. (l97l)» Biochem.

Pharmacol., 20, 463-470.Van Dyke, R.A., Chenoweth, M.B. and Larsen, E.R. (1964a).

Nature (London), 204, 471-472.Van Dyke, R.A., Chenoweth, M.B. and Van Poznak, A. (1964b)

Biochem. Pharmacol., 13, 1239-1247.Victoria, E.J. and Barber, A.A. (1 9 6 9 ). Lipids, 5 8 2 - 5 8 8

Virtue, R.W. and Payne, K.W. (1958). Anesthesiology,19, 562-563.

Vorne, M. and Alavaikko, M. (l97l)» Acta pharmacol. et toxicol., 2 9 , 402-4l6.

Vorne, M. and Arvela, P. (l97l). Acta pharmacol. et toxicol., 29, 417-427. '

Vourc’h, G., Schnoebelen, E., Buck, F. and Fruhling, L. (i960). Anesth. Analg., 17, 466-473*

Walling, C. (1957). Free Radicals in Solution, John .Wiley and Sons, New York.

Ward, P.F.V. and Huskisson, N.S. (1972). Biochem. J., 127, 89-90P.

Wardlaw, J.M., Hennyey, D.J. and Clarke, R.H. (1969). Canad. J. Physiol. Pharmacol., 47, 47-52.

Wills, E.D. (1971)- Biochem. J. , 123, 983-991.Wirtschafter, Z.T. and Cronyn, M.W. (1964). Arch.

Environ. Health, j), 186-191.Wood, J.F., Kennedy, F.S. and Wolfe, R.S. (1968).

Biochemistry, 7., 1707-1713*

Yant, W.P., Schrenk, H.H* and Patty, F.A. (1932). U.S. Bur. Mines Rept. Invest., 31 8 5.

Yllner, S. (l971a )« Acta pharmacol. et toxicol., 29, 471-480.

Yllner, S. (1971b). Acta pharmacol. et toxicol., 29,481-489.

Yllner, S. (1971c). Acta pharmacol. et toxicol., 29, 499-512.

Yllner, S. (l971d). Acta pharmacol. et toxicol., 30, 248-256.

Yllner, S. (l971e)« Acta pharmacol. et toxicol., 30, 257-265.

APPENDIX

Contents.

Table A.l

Table A.2

Table A.3

Table A.4

Reproducibility of peak height ratios in the assay of 1,2-dichloro-1.1.2.2-tetrafluoroethane in tissue extracts.

Tissue distribution of 1,2-dichloro-1.1.2.2-tetrafluoroethane in the Wistar albino rat following an oral dose of 5 6 0 mg/kg body weight.

Tissue levels of trichlorofluoro- methane in unstarved Wistar albino rats after an oral dose of 50 mg/kg body weight.

Full experimental values for tissue levels of trichlorofluoromethane in rats after an oral dose of 50 mg/kg body weight.

Table A.5 Concentrations of trichlorofluoro-methane in blood against time (six female Wistar albino rats).

Table A .1Reproducibility of peak height ratios in the assay of

1 ,2 -dichloro-1 ,1 ,2 ,2 -tetrafluoroethane in tissue extracts.

A 0.327 0.336 0.331

B 0.334 0.338 0.330

C 0 . 1 8 6

0 . 186

0 . 191

D 0.066 0.069

E 0.293 0.307

F 0.056 0.055

G 2.88 2.89

H 4.04 3.96

I 1.52 1.55

J 1.24 1.25

K 0.682 0 . 6 6 0

L 0.692 0.694

M 0.1060 . 105

N 2.442 . 4 7

o 1 . 8 3

1.75

P 1.96 1.92

Q 5.83 5.95

R 0.845 0.844

Tissue

distribution

of 1t

2-di

chlo

ro-l

,1,2,2-tetrafluoroethane

in the

Wistar

albino ra

tan

oral

dose

of 56

0 mg/kg

body

weight

.

VO• ino o rH •CM rH

O + 1 /—n /-N + 1 ^ rH '—' rH '-NCM + 1 VO CM • CM irH LP\ '_■> V—' rH w O ^ + 1 s—'

tH in rH• r- VO

rH VO

inCM n- in• ■H • t CMrH CM o o CM/—s ✓-N /—V /—\ /—'o + 1 ch +1 r - + 1 cn + 1 CM i + 1

VO v_✓ V—' >—' s_x '—'ON cn in O

• • h-CM rH CO •

tO rH CM .0

• -P .dd•rls LA

-ch CM ind • CM CO • ••rl O Oo /—s + 1 /—s + | ^ /—> x~s0 cn +1 in in cn + 1 CM cn CM + 1 ins * cn '_<< ON '— ' '— ' — '•rl cn O CM O oEh • rH • CM

T-1 CM

inin in

• • •o rH rH o CMo *—\ /—s rH '—' /—s /—sCM + i cn + 1 cn cn 1 + 1 CM + 1 cns_/ v_' + 1 w «■—<• '—'

rH o in ON• CO ON •

rH rH CM

incn• • CM

O o •O /-N CM /—- /—s •H t—st-h + 1 cn rH rH + 1 cn 1 1 cn

v-/ >_' — / + 1 V-/rH cn• • inrH CO

0 0H dO toto tod •HS -Po 0

0 cO to3 TJ u d •H 0to 0 0 bO •H ftto 0 > Cl cO u •H•rl rH •rl d U toEh CQ J J ra o <

Continued

over

Table

A*2

(continued)

+ I CMrH

la

LA+ I CA

IAVO

LA

COCO

COCO

LACO VO

+ 1 co

CO

COVO+ 1IA w •st*

+ 1 VO+1CO•H

LALA

CO

Tissue

levels

of trichlorofluoromet

hane

in unstarved

Wistar

albino rats

after

an oral

dose

ofTJ0oH&Hsuo02mm•rl

-P

b O\ftcn

HOO

wcoT5CD000wftX0-pho•H0£*0OJDbOM\bOaoin

CMO

<p O CMCM i i i 1 1 1 1 1 1 • 1 1 o o o 0-0 o • •

o o o

inU20£ in •cP <P CM <P

CM o o CM CM rH rH ’Pi • i i i 1 1 1 1 1 1 • • I • • 1 • • 1 1 1

•H VO o o o o o o

0s

•rl

O CM n f•ip CM cn -cP CM VO envo vo

CM i i i 1 1 1 I I 1 • • t • • • • t • 1 1o o o in cm in in rH CM

O -iP OO VD N N CM -H cP -iP CM -CPVO i i i 1 1 1 1 1 1 • f t • • t • i t 1 1

o o o vo co cn rH CM rH

VO VO CMin t^CO CO O IN rH in cn cn

<p • • • • • • • • • 1 1 1 i l I I I I 1 1o o o in in rH in in o

rH vH

CO VO CM0 o CO 'H VO O O O -iP in vo

-p cn « • t • • • • « • 1 1 1 I I I 1 1 1 1 1

3 o o o in H n CM CM rHs

•Hspi

•rl CO VO • inO n - i n in cv<t< < P c o

0 CM • • i • • • • • 1 1 1 1 I I i I I 1 1 1£ o o O c o rH rH in

•rl CM

o V O o v oO rH O VO rH eP VO CM O N in H eprH • 1 1 1 1 1 • • 1 • • • • • • • • • 1 1

H vH rH ■H O O c o m v o cn in cmH

O CM CO- •

in H e P rH cn O <p ■iP <P CO• • • • • • • • • 1 1 1 I I I I I i 1 1o o o O cn rH O rH CM

m0

0 TJ u H Tl u0 0 0 cO 0 0 bOr—1 0 > •P £ 0 > rt -pCO iH •rl CO 0 rH •H 3

s CQ ft ft CQ A ft

Full

experimental values for

tissue

levels

of tr

ichlorofluorom

ethane

in rats

/-sTJ-. 0• 0

bO rH42

\bO rHs £

o ULA 0<H 00 &

co0 toCO •H0 -PT3

bOH \cd bpu0

to£ cdcd

Tdu oQ) to+> toCh 0cd Uft

X0'_/

orHo

o LA CA + 1CM • •

o o

o

A- CACM rH• • -O o

VO CM VO cf VO vo VOLA CM CM CM O A--H + 1 LA rH CM O VO +1rH • • • • • • rH • • • •rl H H rl O O o o o o COON CM• •

o oLA

A- VOO O• •

VO CO o vo OcO-^ . oA- A- LA VO VO LA-H CAto Is- • • t • + A to O I • • • + 10 o o o o 0 rH o o o o+> CO +> LA

VO ON£ • HP•H o •H ts £ o

A- £ LA•H O •H O• •0 o o o o 0 O -sF co os CO vo VO £ A- LA VO•H LA • • • + 1 •H A- • • • + 1

o o o o o oA-vo vo• •o o

vorH o« •CM CM CM o O -Cf CM orH CO LA CA CM rH

CA • • • • + 1 LA • • • + 1o o o o o o o LACM• •

o o

CAo

VO vo CO O CO COH CM n rH rHrH • • • + 1 CA • •

O O O o ooCMO

• •• •fa fa• •to to+ 1 . 0 + 1

'd rH0 0 d cd ao • rH cd £ cdH id 0 0 0ft S S.- fa S

Continued

over

Table

A.4

(contd.

a - a -o o o o

CO • • CO r l rH r ltH

00

rH

LA• • • LA ■

CA •V£> KD o

o rH tH o -CjH rH + | OCM • • rH ■H CM CM LA VO + |r l

00

LA rH• LA

A - •rH LA

voo

-4ACM VO O IN r lCO O l a a - l a co a - rH LA CO O On r l O n

0 o n • • • + 1 o rH CM + 1 0 VO r l + |•P o o o -p -ppi r l 3 00 VOPI VO Pi rH Pi •

•H • •H • r l CAs o £ £

a p! PI•H LA •H • r l

CM CM<1) t O • 0 LA£ VO O O CO CO o £ CO £ *

•H O CM CACO A -V O •H LA O CA<P ON rH •H LA 4 ^ CM r l A r lH vo • • • t • + | CA CA LA rH + 1 £h -d-1 r l r l r l

O O O O n + 1VOa - • rH

• A - r lo CM

VOCM

• A - ONCM O O LA LA • •

LA CA • • CA O vo vo ON o L A CO CM• • + 1 CM LA CA CA CA r l r l CM r l

o o rH rH + 1 + 1VOVO o

t rH r lo

COo tH

• • A -CO O VO o o o VO N O O •

O ^ CA H v o l a rH • • • LA CM CM CM ON CO CMCA + | O O rH + | rH CM CM CM r l

o o o o o + 1r l CO

• ON• o rH

o

• • •

s s .t • •

(3• • •

m m

0 + 1 + 1 +73 H u0 cd Pi 0 0 ti Pi0 £ «J > h a dt—I <D o •H Ctf 0 0CQ fe s £ S S

Continued

over

Table

A.4

(contdl1

in in• 9o in

o rH CM O +1 o o rH O 41CM rH rH rH CO CM CM CO CM rHo rH rH in• 9

rH inrH rH

• mrH

in co CO o CM CM o ON CM -HrH rH rH CM rH rH CM cn^f

rH rH ino

o.• • 9

CM in 'to o cm r- cM n- •to O on CM -cf1 to O o rH in ON<1) rH rH rH +1 0 ON CM H rl +| 0 VO -cF. cm cn cn 4-i-P -p ■P0 CM 0 CO Pi -tf*0 rH 0 rH Pi cn•H •rl •Hs s sPi 0 Pi•rl n- •rl in •H• * cn0 vo 0 0 9

£ • £ s ON•H in cm +i •H O rH CM +1 •H O rH [s. cn O

cn cm £h VO CM rH Eh cn in CM rH rH +|in

9 • inCO vo CMCM rH

inCM •• • in

cn cnin cn cn tnvo in •H on CM in O ON 4-1

CM rH +| CM rH CM +1 ■H CM incn rH •rH CM rH

CM .•• cnCM

cn CM VO o cn co cnco + l rH 00

rH + I cn rH rH in«orH

» * 9

s s• 9 9fa fa fa• 9 9

m in m0 +i + i 4-1

U H0 0 pi 0 p!> £ cfl -P i—i cd•H 0 0 0 0 ■ <8 ■ 0fa fa S S fa £ S

Continued

over

Table

A.4

(contd.

ON•

03o -d-1 rHON tH 03 03 +1

O03

m in O O ON oCD -4< rH r l rH

-P

•rlas

•rl

CDa•H O c n ^ fE-h cm

m spo cn rH 03 03 o in inrH +> in •

3 ori•rlsa

•rl

0£ CO

VO CN VO •rl oin • t • • £H nH •

o o o o tH

•

•fa

•w

d) +1HRJ g bo 0

+> S «<J ri rHcC CD 0 2 COfa fa g fa s

Table A.5

Concentration of trichlorofluoromethane in blood (/^g/ml)against time (minutes).

Values obtained from blood samples of 6 female Wistaralbino rats after an oral dose of 50 mg/kg.

Time Individual values Mean 4 £>.E.M.(mins) ( xg/ml) (^g/ml)

3 0 . 0 5 8 0 . 0 7 7 o.o4o 0 . 0 2 6 0 . 1 0 0 0 . 0 6 0 + 0.0137 0 . 0 5 6 0 . 2 6 6 0 . 0 7 9 0 . 0 6 2 - 0 . 1 1 6 4 0 . 0 5 0

9 0.249 0.133 0.191 + 0 . 0 5 8

l4-l6 0 . 2 7 8 0 . 2 1 0 0.099 0.295 0 . 2 2 0 + 0.044

19 0.332 0.33224-25 O . 8 1 6 0 . 1 0 5 0 . 5 7 6 0 . 1 3 0 0 . 2 6 2 0.378 + 0 . 1 3 8

40-42 0 . 2 2 2 0.207 0.552 0.313 0.323 4* 0 . 0 8 0

46 0.469 O . 3 6 6 0.417 + 0.0736 i 0.423 0.378 0.238 0.574 0.403 4 0 . 0 7 0

6 5 - 6 6 0 . 1 9 8 0 . 5 2 8 0.363 4 O . 1 6 5

90-93 0.458 0 . 2 0 0 0.284 0.314 + 0 . 0 7 6

9 8 - 1 0 0 0.474 0 . 1 8 0 0.327 4- 0.147120 0.256 0 . 2 5 6

150 0.145 0.124 0.134 4 0 . 0 1 0

l80 0.105 0 . 1 2 8 0.071 0 . 1 0 1 4 0 . 0 1 7