2-chioroethanol formation as evidence for a 2-chioroethyl ... · data were presented for the...

[CANCER RESEARCH 35, 568-576, March 1975]

SUMMARY

Chemical degradation of l-(2-chloroethyl)-3-cyclohexyl-l-nitrosourea or l-(2-chloroethyl)-3-(trans-4-methylcyclohexyl)- 1-nitrosourea in buffer under physiological conditions resulted in the formation of a significant quantity of2-chloroethanol (18 to 25% of the initial nitrosourea concentration). Other degradation products observed includedacetaldehyde (5 to 10%), vinyl chloride (1 to 2%), ethylene (1to 2%), and cyclohexylamine (32%), but not 1,3-dicyclohexylurea. The 2-chloroethyl moiety of l-(2-chloroethyl)-3-cyclohexyl-l-nitrosourea was trapped with halide ions, C1,Br, and I, to form the corresponding dihaloethanes whichwere identified by gas chromatography-mass spectrometrytechniques. High-pressure liquid chromatographic procedures were developed for the separation and quantitation ofthe nitrosoureas and many of their degradation products.

It is postulated that a new mode of l-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea and l-(2-chloroethyl)-3-(trans-4-methylcyclohexyl)- 1-nitrosoureadegradation canoccur thatis not the loss of the chloro group as chloride ion, but theloss of the N-3 hydrogen as a proton. Then the corresponding isocyanate and 2-chloroethyldiazene hydroxide areformed, with the latter intermediate becoming an alkylatingspecies,possibly in part as a 2-chloroethyl carbonium ion.

INTRODUCTION

One of the most promising groups of compounds to bedeveloped by the Chemotherapy Program of the NationalCancer Institute is the 1-(2-chloroethyl)-3-alkyl- 1-nitrosoureas (2). They are chemically reactive compoundsthat decompose nonenzymically at relatively rapid rates inbuffers, even under physiological conditions. Their decomposition rates have been shown to be both pH and iondependent, with BCNU3 displaying a half-life of 43 to 52

I This work was supported by National Cancer Institute Contract

NOl-CM-2320l from the NIH.2 To whom requests for reprints should be addressed.

a The abbreviations used are: BCNU, l,3-bis(2-chloroethyl)-l-

nitrosourea; CCN U , l-(2-chloroethyl)-3-cyclohexyl- l-nitrosourea; methyl

mm at pH 7.2 in phosphate buffer at 37Â°,while CCNU hada half-life of 208 and 64 mm in Tris buffer, pH 7.2, andphosphate buffer, pH 7.2, respectively (2, 10, 11, 13).Montgomery et a!. (13) observed that the decomposition ofBCNU in aqueous media was anomalous, in that decomposition products included acetaldehyde and HC1, rather thanthe corresponding chloroethanol, as expected. They haveproposed that the reaction proceeds via an oxazolidineintermediate formed by cyclization of the ethylene moietyafter solvolysis of the chloro group. These workers statedthat CCNU decomposed in a manner analogous to BCNUto give acetaldehyde, N2, CO2. and cyclohexylamine hydrochloride. However, except for CO2 and N2, no quantitativedata were presented for the decomposition products.

Garrett and Goto (4) have analyzed by gas chromatography isomeric alcohol products formed during the degradation of l-(n-butyl)- l-nitrosourea and l-(tert-butyl)- I-nitrosourea at pH 7.3, and they concluded that carboniumion intermediates were involved in the solvolysis of thesenitrosoureas.

Because of the limited information on CCNU andmethyl-CCNU degradation, the following report describesthe nature of the nonenzymic degradation products and apossible mechanism for their formation. 2-Chloroethanolwas observed as a major product and appeared to be formedvia a mechanism that involves the deprotonation of the3-nitrogen. Solvolysis of the chloro group may also occur toa lesser extent, as reflected by acetaldehyde formation.2-Chloroethyl carbonium ion may participate as an intermediate in CCNU degradation in phosphate buffer.

MATERIALS AND METHODS

Chemicals. CCNU, synthesized by Parke Davis, Lot7503X 105, and methyl-CCNU, synthesized by Merck,

CCNU, l-(2-chloroethyl-3-(trans-4-methylcyclohexyl)-l-nitrosourea; isoCCNU, l-cyclohexyl-3-(2-chloroethyl)-l-nitrosourea; CCU, l-(2-chloroethyl)-3-cyclohexylurea; GC-MS, gas chromatography-mass spectrometry;HPLC, high-pressure liquid chromatography; DNP, 2,4-dinitrophenyl;DNFB, 2,4-dinitrofluorobenzene; DCU, 1,3-biscyclohexylurea.

Received September 18. 1974; accepted November 22, 1974.

568 CANCER RESEARCH VOL. 35

2-Chioroethanol Formation as Evidence for a 2-ChioroethylAlkylating Intermediate during Chemical Degradation of1-(2-Chloroethyl)-3-cyclohexyl- 1-nitrosourea and1-(2-Chloroethyl)-3-( trans-4-methylcyclohexyl)-1-nitrosourea'

D. J. Reed,2 H. E. May, R. B. Boose, K. M. Gregory, and M. A. Beilstein

Department of Biochemistry and Biophysics, Oregon State University, Corvallis, Oregon 97331

on June 27, 2021. © 1975 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

2-Ch!oroethano!from CCNU and Methy!-CCNU Degradation

Sample MFL 619085-00604, were provided by the NationalCancer Institute and shipped from Microbiological Associates, Bethesda, Md. The [chloroethy!-U-14C]CCNU, 9.9mCi/mmole; [cyclohexyl-l-'4C}CCNU, 12.2 mCi/mmole;[2-chloroethyl- U- â€˜4C]methyl-CCNU,5.4 mCi/mmole; and2-[U-'4C]chloroethanol were provided by the NationalCancer Institute.

Acetaldehyde, chloroethane, 2-chloroethanol, 1,3-dicyclohexylurea, 1,2-dichloroethane, I-bromo-2-chloroethane,cyclohexyl isocyanate,and cyclohexylamme were obtainedfrom Eastman Kodak Co., Rochester, N. Y. Dimedone(5 , 5-dimethyl- I ,3-cyclohexanedione), chloroacetaldehydediethyl acetal, 3,5-dinitrobenzoyl chloride, and chloroethylamine hydrochloride were from Aldrich Chemical Co.,Inc., Milwaukee, Wis.

All solvents were J. T. Baker reagent grade. Ethylene,ethane, propane, and vinyl chloride were commerciallyavailable as compressed gases. Porapak Q 80-100 mesh wasobtained from Walters Associates, Inc., Framingham,M ass.

Nitrosourea Syntheses. 1,3-Bis(cyclohexyl)- 1-nitrosoureawas synthesized by anhydrous nitrosation of 1,3-dicyclohexylurea according to the method of Johnston et a!. (7, 8).iso-CCNU was synthesized as an isomeric mixture (about70:30, CCNU:iso-CCNU) by nitrosation of CCU in thepresence of water according to the procedure ofJohnston eta!. (8). Proof of the formation of iso-CCNU was thedistribution of â€˜4Cbetween the isomeric pair after theirsynthesisfrom [ch!oroethy!-U-'4C]CCU followed by liquidchromatographic separation. The â€˜4C-labeled CCU wassynthesized by the reaction of [ch!oroethy!-U-'4C]BCNUwith cyclohexylamine, according to the method of Montgomery et a!. (13).

Gas Chromatography. Reaction mixtures were preparedin l0-ml Erlenmeyer flasks equipped with serum vialstoppers. Head spacewas 10.0 ml, and gassamplesup to 1.0ml were removed for gas chromatographic analysis. Knowngas volumes of propane were used as an internal standardfor the reactions in which the evolution of volatile productsfrom CCNU degradation were measured by gas chromatography. A Model 810 F&M gas chromatograph equippedwith a dual flame detector was used with a Â½-inchx 4-ftstainlesssteelcolumn packed with Porapak Q 80-100 mesh.

GC-MS. Electron impact mass spectra were obtainedwith a Varian Model M7 gas chromatograph-mass spectrometer system at an ionizing potential of 70 V. The samePorapak Q column was used when sample introduction intothe mass spectrometer was via the gas chromatograph.

Selective Solvent Extraction of [â€˜4C]CCNU IncubationMixtures. Reaction mixtures were 1stextracted 3 times withhexane. This procedure extracted essentially all of theunreacted CCNU but very little (5%) of the polar products.The reaction mixtures were then extracted 3 times withdiethyl ether (ether) which extracted polar products such as2-chloroethanol. Extracted incubates were radioassayed fornonextractable products. The hexane and ether extractswere dried over anhydrous sodium sulfate and maintained inthe dark and cold prior to analysis by liquid chromatography.

Liquid Chromatography. HPLC was conducted with aModel 3122 chromatograph equipped with a Model HPSVsample injection valve, a 20- or 100-zl sample loop, a Model200 UV photometer (Chromatronic, Inc., Berkeley, Calif.),and a Model 52 111 Omniscribe strip chart recorder (Houston Instrument Div., Bellaire, Texas). Certain compounds,which do not possess UV absorbance at 254 nm, weresynthesized from â€˜4C-labeled intermediates and characterized by liquid chromatography and liquid scintillationcounting ofO.5-ml eluate fractions (Table 1). This approachprovided a convenient method for quantitation of 2-chloroethanol formed during the degradation of CCNU andmethyl-CCNU as described below. The values obtainedwere further verified by reverse isotope dilution analyses ofthe 3,5-dinitrobenzoate derivative of 2-chloroethanol.

HPLC of Nitrosoureas. HPLC with a 2. 1- x 500-mmcolumn of Bio-Sil A (20- to 44-@zm mesh size, Bio-RadLaboratories, Richmond, Calif.) was used to quantitateCCNU, iso-CCNU, and methyl-CCNU. Detection was bya 254-nm UV detector. Following loop injection of 20-alaliquots of a hexane solution of nitrosourea, the column wasisocratically eluted with isooctane:chloroform, 5: 1 (v/v), at500 psig and a flow rate of I .5 ml/min. Retention times formethyl-CCNU, CCNU, and iso-CCNU were 5.5, 6.0, and9.0 mm, respectively. Detection response was linear from0.1 to 50 @g.

Reverse Isotope Qilution Analysis for 2-[U-â€˜4C]Chloroethanol.After hexane extraction of [14C}CCNUreaction mixtures, as indicated above, 500 mg of 2-chloroethanol were added before ether extraction. The combinedether extracts were evaporated carefully to remove theether. The 3,5-dinitrobenzoate derivative was formed according to the procedure of YlIner (19) and recrystallizedfrom 100% ethanol until a constant specific activity wasobtained, as determined by liquid scintillation counting. Inother experiments, the 3,5-dinitrobenzoate derivative wasformed without carrier 2-chloroethanol and analyzed byH PLC.

Reverse Isotope Dilution Analysis for â€˜4C-Acetaldehyde.Incubations of [ch!oroethyl-'4CJCCNU in 0. 1 M phosphatebuffer, pH 7.4, were conducted in Erlenmeyer flasks sealedwith rubber serum vial stoppers. At the end of the incubation, a known quantity of carrier acetaldehyde was injectedinto the reaction mixture and mixed well. A nitrogen gasstream was then used to sweep the [U-14C]acetaldehyde intoa trap cooled by Dry Ice and containing dimedone. Themethone derivative of acetaldehyde was prepared accordingto the procedure of Horning and Horning (5) and recrystallized to constant specific activity, which usually required 3to 5 recrystallizations.

Reaction of 2-Chloroethylamine with Nitrous Acid. 2-Chloroethylamine hydrochloride (1 mmole) was dissolved inglass-distilled water (5 ml) contained in a 10-mi Erlenmeyerflask sealed with a rubber serum vial stopper. Sodium nitrite(3 mmoles) in water (1 ml) was injected into the flask, andthe reaction mixture was incubated at 50Â° in a shakingwater bath for 2 hr. Both gaseous and liquid aliquots wereremoved for analysis. The liquid aliquots were mixed withn-octanol (0. 1 ml) and the n-octanol phase was assayed for

MARCH 1975 569



Isooctane:â€˜4C-Labeled

compoundColumn (2. 1 x 500 mm)chloroformaratio (v/v)Elutionrate

(mi/mm)Retentiontime

(mm)2-ChloroethanolBio-Sil

A2:11.022-262-ChloroethanolBio-SilA4:51.215-182-Chloroethyl-3,5-Bio-Sil

A5:12.08-9dinitrobenzoate2-Chloroethyl-N-cyclo

Vydac2: 11.232-35benzylcarbamateCCUVydac2:11.09-11CCUBio-Sil

A2: 11.0>60DCUVydac2:11.05-7

D. J. Reed et a!.

Table I

Liquid chromatography conditions for some CCNU-related compounds

a Eluting solvent.

2-chloroethanol by gas liquid chromatography on a Â½-inchx 6-ft stainless steel column packed with 14% (w/w)diethylene glycol succinate adsorbed on Gas Chrom Q. Thegaseous aliquots were analyzed on a Porapak Q column.

DNP Amine Derivative Preparation. After incubation ofeither BCNU or CCNU in 0. 1 M phosphate buffer, pH 7.4,the remaining unreacted nitrosourea was removed by hexane extraction. Next, the more polar degradation productssuch as 2-chloroethanol were removed by ether extraction.Then 3 ml of acetone, 25 zl of DNFB, and 50 mg ofNaHCO3 were added to the extracted incubation mixture(2.0 ml). The reaction mixture was then warmed in a cappedtube at 60Â°for 30 mm. Upon cooling, 25 mg ofglycine wereadded and the mixture was heated again to eliminate anyunreacted DNFB. After cooling, the mixture was extracted3 times with ether, and the ether extracts were combined,dried with anhydrous sodium sulfate, and evaporated todryness with a stream of dry nitrogen gas. It was possible toobtain a quantitative yield of DNP derivatives of primaryand secondary amines in either buffer or tissue homogenateswith a minimum of interfering substances. At pH 8.5 to 9.0,ether extraction of either the amino acid derivatives or 2,4-dinitrophenol is prevented while the DNP amine derivatives are extracted quantitatively with 3 extractions.

Preparation of Known DNP Derivatives. Equal volumesof DNFB and amine as the free base were reacted asdescribed above and recrystallized from 95% ethanol. Whenamine hydrochloride salts were used, the amine hydrochloride was converted to the free amine with sodium carbonateprior to the addition of DNFB.

HPLC of DNP Amine Derivatives. Samples were injectedvia a IOO-zl sample loop onto a 2. 1- x 500-mm Bio-Sil Acolumn at 500 psi. The column was eluted isochraticallywith isooctane:chloroform, 2: 1, at a flow rate of I .4 ml/min.The detector was a UV monitor at 254 nm. The retentiontimes for cyclohexyl-, n-propyl-, ethyl-, methyl-, and 2-chloroethyl-amine DNP were 3.3, 4.4, 5.4, 9.0, and I 1.5,respectively. Essentially baseline separation was achievedwith a linear detector response from 0. 1 to 10 @gof eachDNP amine derivative.

RESULTS

Decomposition Rate of CCNU and Methyl-CCNU. Thedecomposition of CCNU and methyl-CCNU at 37Â°in 0. 1 M

phosphate buffer, pH 7.4, was examined by a time course ofnitrosourea disappearance, using hexane extraction followed by analytical HPLC. The degradation of CCNU ormethyl-CCNU was a pseudo 1st-order rate process at pH7.4 with a half-life of 48 Â±4 and 70 Â±7 mm, respectively(Chart 1).

Artifactual Formation of CCU. Extraction of CCNUfrom phosphate buffer with hexane caused the appearanceof a polar compound in the hexane extract, but not in theincubation mixture. The concentration ofthis substance was10 to 15% of that of the extracted CCNU. Liquid chromatography followed by isolation and mass spectrometry ofthis substance gave data that indicated the formation ofCCU (Table 2). The CCU present in the hexane extractsgave a mass spectral pattern that is entirely different fromCCNU (12) or 2-chloroethyl-N-cyclohexyl carbamate but isin good agreement with synthetic CCU, which was preparedby reaction of 2-chloroethyl isocyanate with a slight excessofcyclohexylamine (13). Using [â€˜4C]CCU, it was found thathexane would only partially extract CCU from 0. 1 Mphosphate buffer, pH 7.4. Yet, no CCU was found inhexane-extracted incubates of CCNU in buffer. Therefore,

Chart I. Decomposition with time of CCNU or methyl-CCNU in 0.1M phosphate (P04) buffer, pH 7.4. The nitrosourea (1.8 @mo1es)was

dissolved in 200 @lof acetone and added to 3.0 ml of buffer. The reactionmixture was incubated in the dark with shaking at 37Â°. Nitrosoureaconcentration was determined by hexane extraction and HPLC on Bio-SilA, as described in the text.

TIME ( MIN)




Retention time (mm)withgaschromatographâ€•Compoundcolumn

temperatureQuantityformed

(nmoles/ 180mm)100Â° 175Â°Ethylene0.5518-36Ethane0.71

Quantity

Nitrosourea TimeIncubation (l800nmoles) Amine formed (mm)nmoles0.

1 M phosphate buffer, pH 7.4BCNU

2-Chloroethylamine 30 270CCNU 2-Chloroethylamine 30 30

(30 160@ 60 220CCNU Cyclohexylamine@ 120 380

(180 580

D. J. Reed Ct a!.

represented 20 to 25% of the total â€˜4Cadministered as[ch!oroethy!-'4C]CCNU. The incubation medium containedconsiderable â€˜4Cthat was not hexane or ether extractable.Reaction of the extracted incubates with l-fluoro-2,4-dinitrobenzene did not yield an appreciable amount (

2-Ch!oroethano!from CCNU and Methy!-CCNU Degradation

2-chloroethanol formation and promoted the formation ofl,2-dichloroethane from 7 to 25 nmoles/30 mm. Also, whenTris buffer was used, 2-chloroethanol formation was considerably greater in the absence (205 nmoles/30 mm) than inthe presence of 4 M KCI (127 nmoles/30 mm).

Bromide ion is a better nucleophile than chloride ion (16)and would be expected, on the basis of an 5N2 typemechanism, to undergo reaction more rapidly than chlorideion if the participating intermediate(s) do not lead to theformation of a carbonium ion. If the reaction is simply anionic interaction via an 5N I type mechanism (as shown inChart 6), then little preference would be shown in thereaction of 2-chloroethyl carbonium ions with competingchloride and bromide ions. Incubations of CCNU in 0. 1 Mphosphate buffer, pH 7.4, in the presence of 2 M KCI and 2 MKBr resulted in the formation of approximately equalquantities of I ,2-dichloroethane and l-bromo-2-chloroethane over a period of 3 hr and therefore supports an 5N 1mechanism more than an 5N2 mechanism (Table 7). Thetotal yield ofdihaloethanes in 180 mm represents about 15%of the CCNU undergoing degradation. Incubations with KIpresent resulted in the formation of l-chloro-2-iodoethane.Identity of the dihaloethanes was established by GC-MS(Table 8).

DISCUSSION

In order to understand better the mechanism of action ofnitrosoureas as carcinostatic agents, efforts have beendirected toward a detailed investigation of their chemicaldegradation. Attention has been given to both the natureand quantities of products formed, particularly those im

DEGRADATION OF 2-CHLOROETHYLNITROSOUREAS

CICH2CH2â€”@Nâ€”Ã´H

CICH2CH?+N2 4@OH

@/C@HiCH2@@*@CICH2CH2X CICH2CH2SR CICH2CH2OPO3H

Chart 6. Chemical degradation scheme proposed for nitrosoureas. Theinitial step is a base-catalyzed formation of a proton leading to thehypothetical 2-chloroethyl diazene hydroxide which immediately decomposes in a manner that leads to 2-chloroethanol formation as well as otherdegradation products.

TIME (MIN)

LI

.J4I-0

0

2ILlU

w0.

Chart 5. Decomposition of [carbonyl-â€•C]CCNU in 0. 1 M phosphatebuffer, pH 7.4, at 37Â°.Details are the same as those in Chart 2.

incubates was extracted by hexane extraction and theremaining two-thirds, by ether extraction.

Degradation Products of 2-Chloroethylamine Nitrosation.If a major portion of CCNU and methyl-CCNU was beingdegraded via the deprotonation of the N-3 nitrogen to formthe corresponding isocyanate and a nitrosated 2-chloroethylamine moiety, then the degradation of N-nitroso-2-chloroethylamine could be expected to yield similar products asthe 2-chloroethyl moiety of CCNU and methyl-CCNU(Chart 6). Experiments in which 2-chloroethylamine hydrochloride was nitrosated in an aqueous solution by sodiumnitrite resulted in the formation of 2-chloroethanol, vinylchloride, and ethylene in yields which compared in magnitude to those observed during CCNU degradation in phosphate buffer, pH 7.4.

Halide Ions as Nucleophilic Trapping Agents. The proposed chemical breakdown of CCNU involves the intermediate formation of 2-chloroethyl diazene hydroxide (Chart6). Formation of 2-chloroethanol from this compound canbe explained by a nucleophilic displacement of the diazenehydroxide group by a hydroxyl ion (or water) (SN2), or byspontaneous decomposition of 2-chloroethyl diazene hydroxide to form 2-chloroethyl carbonium ion and hydroxylion followed by ionic interaction of these 2 groups (SN 1). Tounderstand in greater detail a possible mechanism ofdecomposition relative to 2-chloroethanol formation, halideions were used as nucleophiles. Since halide ions havedifferent nucleophilicities, each should compete differentlywith hydroxyl ion or water to produce different ratios ofdihaloethane to 2-chloroethanol if an 5N2 mechanism isoperative. On the other hand, if an 5N I mechanism isoperative, the halide ions may compete with the â€œcagedâ€•hydroxyl ion to trap the carbonium ions and form thecorresponding dihaloethanes. Experiments were conductedin an effort to distinguish between the 5N2 and 5N Imechanism using either phosphate or Tris buffers in thepresence of 4 M KCI (Tables 6 and 7). The addition of 4 MKCI did not significantly change the rates of CCNUdegradation or 2-chloroethanol formation in phosphatebuffer. However, the presence of 4 M chloride ion caused 37nmoles of l,2-dichloroethane to be formed in 30 mm (Table6). The addition of 4 M KC1 to 0. 1 M Tris-chloride buffersignificantly decreased the rates of CCNU degradation and

CICH2CH2NCONH-R R = â€”CH2CH2CI BCNU

j;N@O : -Q CCNU

CICH2CH2N1C@Nâ€”R : trans

N HH k

CICH2CH2N + OCNâ€”R@SOCYANATE)

N-OH + OW

gH3

METHYL CCNU

OCN-R

H2O@

HOOCNHâ€”RCICH=CH2

@-cO2

H2Nâ€”R

OCN-R

RNHCONHR

MARCH 1975 573



I ,2-Dichlo 2-ChloroCCNUroethaneethanoldegradedformedformed(nmoles/30(nmoles/30(nmoles/30Conditionsmm)mm)mm)0.1

M phosphate buffer,521 Â±471139 Â±17pH7.40.1

M phosphate buffer,455 Â±5337 Â±I138 Â±35pH7.4, plus 4 MKCI0.lMTris-chloride403Â±

187Â±I205Â±20buffer,pH7.40.1

MTris-chloride335 Â± I25127 Â±13buffer.pH 7.4,plus4M

KCI

Incubation time(mm)Dihaloethane

formed(nmoles)CICH2CH2CICICH2CH2Br15243030464260697490829312097109180121139

m/eStructure assignmentRelative

intensitiesStandards

DCEbBCESamplesâ€•I2327

49516263646598

100107109142144155190192CH2=CH@

35CI@=CH237CI@=CH235Cl@â€”CH=CH235CIâ€”CH,-â€”.CH2@37C1@â€”CH=CH,37CIâ€”CH@----CH2@35Clâ€”CH@â€”â€”CH@â€”-35Cl@35C1â€”CH@â€”CH@-â€”-37Cl@

79Brâ€”CH@,â€”.CH2@â€œBrâ€”CH@,â€”--CH2@79Br@â€”CH@-â€”.CH@-â€”35Clâ€œâ€˜Br@â€”CH@--â€”CH@-â€”35Cl

Iâ€”CH@--â€”CH,@l@â€”CH@----CH,---35Cll@â€”CH@-â€”.CH@-â€”37Cl190

55725

100017730380

215177518

191

1000

667

2045685

107230

372

1000385436282385333315

1000

320

93102157176625

1000

452

174521191

D. J. Reed et a!.

mediates which may explain the nature of the pharmacodynamically important cellular alkylation and/or carbamoylation reactions.

Evidence has been presented that demonstrates the formation of 2-chloroethanol as well as acetaldehyde, vinylchloride, ethylene, and cyclohexylamine as products ofdegradation of CCNU in phosphate buffer, pH 7.4. Theextent of 2-chloroethanol formation, 20 to 25% of totalCCNU in 3 hr at 37Â°,which is 3 to 4 times greater thanacetaldehyde formation, raises doubt as to whether themechanism of decomposition presented by Montgomery eta!. (13) for BCNU and CCNU is valid for CCNU and

Table 6

Formation of l,2-dichloroethane during CCNU degradation in thepresence of chloride ion

CCNU, 1800 nmoles, was dissolved in 200@ of absolute ethanol andadded to 3.0 ml of the specified buffer and KCI solution. The reactionmixtures were then analyzed by GC-MS as described in â€œMaterials andMethods.â€•

methyl-CCNU. Their (13) mechanism, in contrast to thescheme presented in Chart 6, involves an initial loss ofchloride ion to form an unstable oxazolidine intermediatethat rapidly yields ethylenediazenehydroxide and the alkylisocyanate corresponding to the 3-alkyl moiety of the1-nitrosourea. Montgomery et a!. (13) stated that the modeof decomposition of CCNU was analogous to BCNU togive acetaldehyde, N2, CO2. and cyclohexylamine hydrochloride. Our data support a reaction mechanism thatrequires base abstraction of the N-3 hydrogen, resulting inan unstable intermediate that rapidly decomposes to yieldthe corresponding alkyl isocyanate and 2-chloroethyl diazene hydroxide rather than ethylene diazene hydroxide.Subsequent spontaneous decomposition of 2-chloroethyldiazene hydroxide could yield a 2-chloroethyl carboniumion rather than a vinyl carbonium ion (17).

Evidence supporting a degradation mechanism involvinghypothetical formation of a 2-chloroethyl diazene hydroxide

Table8

Table 7

Simultaneous formation of 1-bromo-2-chloroethane and1,2-dichloroethane during the degradation of CCNU in the presence of

bromide and chloride ions

The reaction conditions were the same as those described in Table 6,except that KC1 and KBr were 2 Meach in 0. 1 M phosphate buffer, pH 7.4.

Mass spectral identification of l-halo-2-chloroethanes formed during degradation of CCNU inphosphate buffer in the presence of 4 M KX

a Samples derived from gaseous atmosphere after incubation of CCNU in the presence of 4 M

KX, where X = Cl for Sample 1, Br for Sample 2, and I for Sample 3.a DC E, dichloroethane; BCE, l-bromo-2-chloroethane.

CANCER RESEARCH VOL. 35574



2-Ch!oroethano!from CCNU and Methyl-CCNU Degradation

intermediate rather than an ethylene diazene hydroxideintermediate has been obtained from (a) the substantialyield of 2-chloroethanol (20 to 25%) and (b) the ability touse halide ions as nucleophiles to trap the 2-chloroethylmoiety as the corresponding l,2-dihaloethanes. The extentthat the I ,2-dihaloethanes were formed corresponded wellwith the reportedly weak alkylating activity ofCCNU using4-(p-nitrobenzyl)pyridine (18). The low yield of acetaldehyde is taken as evidence that ethylene diazene hydroxidemay be a less important intermediate. 2-Chloroethanolprobably does not arise from a nucleophilic attack at C-I ofthe chloroethyl group by hydroxyl ion (SN2) becausenucleophiles in high concentration do not increase and, insome cases, decrease the rate of decomposition of CCNU.Furthermore, CCNU is relatively stable in buffered solutions at low pH.

Our data support an 5N 1 breakdown of 2-chloroethyldiazene hydroxide and not an 5N2 mechanism. The proposed formation of the 2-chloroethyl carbonium ion is in(consistent) agreement with the results of Garrett and Goto(4) who have described the kinetics of solvolysis of variousl,3-dialkyl-N-nitrosoureas in neutral and alkaline solutions.They concluded that degradation of l,3-disubstituted nitrosoureas was subject to spontaneous and specific hydroxide ion catalysis. They examined the alcohol products fromalkaline degradation of l-(n-butyl)- 1-nitrosourea and I-(tert-butyl)- I -nitrosourea and found alkyl rearrangementsthat implicated carbonium ion participation in alcoholformation.

1-(2-Chloroethyl)-3-alkyl- 1-nitrosoureas are sufficientlyreactive under physiological conditions to suggest a therapeutic efficacy which does not require enzyme-catalyzedtransformation. May et a!. (12) have observed a rapidmicrosomal hydroxylation of the cyclohexyl moiety ofCCNU. Ring hydroxylation does not alter the proposeddegradation scheme (Chart 6) but does enhance the polarityof the drug and could modify the rates of formation ofalkylating and carbamoylation species. Studies of thechemical stability of these hydroxylated products are currently in progress.

What are the implications of a 2-chloroethyl alkylatingmoiety in contrast to a vinyl moiety? Much remains to beunderstood in this regard. Dr. D. J. Reed and coworkershave observed, in work to be reported elsewhere, thatessentiallyall ofthe ethylene moiety ofCCNU that appearsin the urine of rats administered CCNU is present ascompounds such as thiodiacetic acid derived from sulfurconjugation. Similar conjugates have been reported byYlIner (19), who has examined the metabolism of 2carbon-halogenated compounds. Johnson (6) studied themetabolism of 2-chloroethanol in rats and found liverglutathione to be rapidly depleted with the concomitantformation of S-carboxymethyl glutathione, which is furthermetabolized prior to excretion. Importantly, 2-chloroethanol is converted to 2-chloroacetaldehyde by alcohol dehydrogenase (6), the product being a rapid nonenzymicalkylating agent of free thiol groups. Lawrence et a!. (9)have reported that 2-chloroacetaldehyde is about 10 to 30times as toxic as 2-chloroethanol in vivo in rats. However,

when tested with L-cells in culture, chloroacetaldehyde,when present at 5.6 x l0@ M, was more than 1200 times astoxic as 2-chloroethanol and could decrease the rate ofprotein synthesis by 50%. Thus, both the alkylating abilityof the 2-chloroethyl moiety (as a carbonium ion or a2-chloroethyl-substituted nucleophile such as phosphate), aswell as the formation of 2-chloroethanol and its potentialalkylating ability via 2-chloroacetaldehyde, could play animportant role in the overall pharmacological response toCCNU and methyl-CCNU. Alkylation of cellular constituents by a 2-chloroethyl carbonium ion could result in thepotential for ethylene cross-linkages via alkylation, concomitant with the chloro group being displaced by amacromolecular nucleophile such as a protein thiol group.

in vitro carbamoylation of the c-amino group of lysine ofproteins by the spontaneousdegradation ofCCNU has beenreported (3, 15). Previously, Montgomery et a!. (13) reported the formation of cyclohexylamine hydrochlorideduring CCNU degradation. Oliverio et a!. (14) detectedradioactive cyclohexylamine (14 to 18% of radioactive dose)in the urine of mice administered CCNU labeled with â€˜4Cinthe cyclohexyl moiety. CCNU degradation in phosphatebuffer, pH 7.4, results in the formation of cyclohexylaminewith yields up to 32% of the CCNU present in 3 hr at 37Â°.Surprisingly, the yield of l,3-bis-cyclohexylurea after 3 hrwas less than 0.3% of the original CCNU present. Themarked instability of alkyl isocyanates has been noted byother workers and could possibly explain the lack offormation of the corresponding urea by a reaction betweencyclohexylamine and cyclohexylisocyanate. For example,Brown and Wold (I) determined that the half-lives of octyland butyl isocyanates were approximately I mm in 0.08 MTris-HCI buffer, pH 7.7, containing 0. 1 M CaC12. However,they found these isocyanates to be active site-specificreagents for proteases in vitro. Such reactivity of isocyanates emphasizes the need for a better understanding of thepossible pharmacological roles of the isocyanate moieties aswell as the 2-chloroethyl moiety resulting from the degradation of the nitrosoureas.

ACKNOWLEDGMENTS

We are indebted to Dr. Robert R. Engle and the Cancer ChemotherapyNational Service Center of the National Cancer Institute for providing theâ€œC-labeled preparations of CCNU. methyl-CCNU. and 2-chloroethanol.

REFERENCES

1. Brown, W. E., and Wold, F. Alkyl Isocyanates as Active-Site-SpecificReagents for Serine Proteases. Reaction Properties. Biochemistry. 12:828-834, 1973.

2. Carter, S. K., Schabel, F. M., Jr., Broder. L. E., and Johnston, T. P.l,3-Bis(2-chloroethyl)-l-nitrosourea (BCNU) and Other Nitrosoureasin Cancer Treatment: A Review. Advan. Cancer Res., 16: 273-332,1972.

3. Cheng, C. J., Fujimura, S., Grunberger. D.. and Weinstein, I. B.Interaction of l-(2-Chloroethyl)-3-cyclohexyl-l-nitrosourea (NSC79037) with Nucleic Acids and Proteins in Vivo and in Vitro. CancerRes.,32:22-27,1972.

4. Garrett, E. R., and Goto, S. Kinetics of Solvolyses of Various

MARCH 1975 575



D. J. Reed et a!.

N,N â€˜-Dialkyl-N-nitrosoureas in Neutral and Alkaline Solutions.Chem. Pharm. Bull. Tokyo. 21: 1811-1823, 1973.

5. Horning, E. C., and Horning, M. G. Methone Derivatives ofAldehydes.J.Org. Chem., 1!.95-99, 1946.

6. Johnson, M. K. Metabolism of Chloroethanol in the Rat. Biochem.Pharmacol., 16: 185-199, 1967.

7. Johnston, T. P.. McCaleb, G. S., and Montgomery, J. A. TheSynthesis of Antineoplastic Agents. XXXII. N-Nitrosoureas. I. J.Med.Chem.,6:669@68l, 1963.

8. Johnston, T. P., McCaleb, G. S., Opliger, P. S., and Montgomery, J.A. The Synthesis of Potential Anticancer Agents. XXXVI. N-Nitrosoureas. II. Haloalkyl Derivatives. J. Med. Chem., 9: 892-911,1966.

9. Lawrence, W. H., Dillingham, E. 0., Turner, J. E., and Autian, J.Toxicity Profile of Chloroacetaldehyde. J. Pharm. Sci., 61: 19-25,1972.

10. Loo, T. L.. and Dion, R. L. Colorimetric Method for the Determination of l,3-bis(2-Chloroethyl)-l-nitrosourea. J. Pharm. Sci., 54:809-810.1965.

II. Loo, T. L., Dion, R. L., Dixon, R. L., and Rail, D. P. The AntitumorAgent, l,3-bis(2-Chloroethyl)-l-nitrosourea. J. Pharm. Sci., 55.487-492. 1966.

12. May. H. E., Boose, R., and Reed, D. J. Hydroxylation of theCarcinostatic l-(2-Chloroethyl)-3-cyclohexyl- 1-nitrosourea (CCNU)by Rat Liver Microsomes. Biochem. Biophys. Res. Commun., 57:

426-433, 1974.13. Montgomery, J. A., James, R., McCaleb, G. S., and Johnston, T. P.

The Modes of Decomposition of l,3-Bis(2-chloroethyl)-l-nitrosoureaand Related Compounds. J. Med. Chem., 10: 668-674, 1967.

14. Oliverio, V. T., Vietzke, W. M., Williams, M. K., and Adamson, R.H. The Absorption, Distribution, Excretion, and Biotransformation ofthe Carcinostatic l-(2-Chloroethyl)-3-cyclohexyl- 1-nitrosourea in Animals. Cancer Res., 30: 1330-1337, 1970.

15. Schmall, B., Cheng, C. J., Fujimura, S., Gersten, N., Grunberger, D.,and Weinstein, I. B. Modification of Proteins by l-(2-Chloroethyl)-3-cyclohexyl-l-nitrosourea (NSC 79037) in Vitro. Cancer Res., 33:1921-1924,1973.

16. Swain, C. G., and Scott, C. B. Quantitative Correlation of RelativeRates. Comparison of Hydroxide Ion with Other Nucleophilic Rcagents toward Alkyl Halides, Esters, Epoxides and Acyl Halides. J.Am. Chem. Soc.,75:141-147,1953.

17. Wheeler, G. P., Bowdon, B. J., Grimsley, J. A., and Lloyd, H. H.Interrelationships of Some Chemical, Physiocochemical, and Biological Activities of Several l-(2-Haloethyl)-l-nitrosureas. Cancer Res.,34: 194-200, 1974.

18. Wheeler, G. P., and Chumley, S. Alkylating Activity of l,3-Bis(2-chloroethyl)-l-nitrosourea and Related Compounds. J. Med. Chem.,10:259-261,1967.

19. YlIner, S. Metabolism of l,2-Dichloroethane-â€•C in the Mouse. ActaPharmacol. Toxicol., 30: 257-265, 1971.




1975;35:568-576. Cancer Res D. J. Reed, H. E. May, R. B. Boose, et al.

-4-methylcyclohexyl)-1-nitrosoureatrans1-(2-Chloroethyl)-3-(1-(2-Chloroethyl)-3-cyclohexyl-1-nitrosourea andAlkylating Intermediate during Chemical Degradation of 2-Chloroethanol Formation as Evidence for a 2-Chloroethyl

Updated version

http://cancerres.aacrjournals.org/content/35/3/568

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/35/3/568To request permission to re-use all or part of this article, use this link


http://cancerres.aacrjournals.org/content/35/3/568http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/35/3/568http://cancerres.aacrjournals.org/

2-chioroethanol formation as evidence for a 2-chioroethyl ... · data were presented for the...

Documents