In the Search for New Anticancer Drugs. XXIII: Exploration of a Predictive Design for Anticancer Drugs of Carbohydrates Containing N-Nitrosochloroethylamino, N-Nitrosomethyl, and N-Nitrosoaminoxyl Components

In the Search for New Anticancer Drugs. XXIII: Exploration of a Predictive Design for Anticancer Drugs of Carbohydrates Containing MNitrosochloroethylamino, N-Nitrosomethyl, and N-Nitrosoaminoxyl Components GEORGESOSNOVSKY'

AND

NUTI UMA MAHESHWARA RAO

Received May 25, 1990, from the Department of Chemistry, University of Wisconsin-Milwaukee, Milwaukee, WI 53201. publication October 1, 1990.

Accepted for

~.

Abstract The spin-labeled glucose nitrosoureas 13-16, streptozotocin (l8),chlorozotocin (31),streptozotocin analogues of galactosyl24 and mannosyl28, and their tetra-Oacetylderivatives 25 and 29, MCNU (Cymerin, 34), and glucamine (21) were synthesized and evaluated in vivo for their anticancer activities against the murine lymphocytic leukemia P388. Compounds 13-16, 18, 24, 28, 31, and 34 possessed activities ranging from 33 to 603% increase in life span (%ILS),whereas 21, 25, and 29 were inactive (%ILS = 9 to 10). All CD,F, male mice treated with the most active compounds (13, 14, and 34) at 20 mglkg were alive after 30 days, whereas all mice treated with the clinical drug streptozotocin (18) and clinically tested chlorozotocin (31) succumbed. Compounds 13-16, 18, 31, and 34 were further evaluated for their antineoplastic activity against lymphoid leukemia L1210. Compounds 13 and 34 on day 60 exhibited ?!OILSvalues of 557 and 713, respectively, as compared with %ILS values of 646 and 713 for CCNU (1) and the spin-labeled SLCNU (3),respectively.The lipophilicitiesof 13-16,18,21, 24, 25, 28, 29, 31, and 34 were determined using EPR and/or UV methods. A predictive design pattern was observed, with the most active drug (34) possessing some hydrophobic property (log P = 1.24), followed by 13 (log P = 1.87)and 14 (log P = 1.81) as the more active drugs with higher hydrophobicity than 34. The clinical drugs streptozotocin (18) and chlorozotocin (31) were distinctly hydrophilic and less active. Finally, it was concluded that various scattered results of anticancer activity in the literature can be explained by a linear correlation of activities with lipophilicities.

3. R

4.R-

-

+jL0

fi 0

Nitrosoureas

was found22-25 that 3 is less toxic and more active than the clinical drugs 1 and 2, as evidenced by the therapeutic index for 3 of 40 as compared with those for 1 and 2 of 5-7.25 It was Alkylnitrosoureas (see structure), particularly the 2-chloshown that while the aminoxyl moiety imparts a beneficial roethylnitrosoureas, such as N'-cyclohexyl-N-(2-chloroethyl)- influence on the antineoplastic activity of a drug,22-25 the N-nitrosourea (CCNIJ, 1) and N'-(truns-4-methylcyclohexyl)- aminoxyl radical by itself has no anticancer activity,28 is not mutagenic nor carcinogenic,29 is relatively nontoxic,28 exhibN-(2-cholorethy1)-N-nitrosourea(MeCCNU, 2), have been its no synergetic effect,30 and has only little effect on the cell extensively investigated in vitro and in vivo in a wide variety growth and cell cycle kinetics.31 Further, it was found32 that of human and animal neoplasms. 1-27 The structureactivity relationship of this class of compounds has been extensively many aminoxyl radicals readily migrate through cell memstudied.4.7-11 In spite of these studies, the mechanisms of the branes into the intracellular space, and can even cross the anticancer action and the causes of toxicity of these drugs blood-brain barrier32 at sites of diseased tissues. On the basis under physiological conditions are not fully understood.t?-14 It of these results, it was hypothesized25933 that the aminoxyl is assumed11-14 that the monofunctional nitroso and difuncradical is a carrier moiety that facilitates the transport of the tional chloroethylnit,rosoureas spontaneously decompose in drug through biological membranes on its way to the cellular DNA. vivo, probably without biological activation, to give a variety of derivatives which undergo either electrophilic alkylation Further detoxification of these drugs can be achieved by the and/or interstrand cross-linking of DNA and of other biomolreplacement of the cyclohexyl groups in 1 and 2 with various ecules, or carbamoylation reactions of amino groups in biocarbohydrate moieties. This approach led to the development logical systems. Although nitrosoureas such as 1 and 2 exhibit of several carbohydrate derivatives3a4 of nitroso- and chloa wide variety of severe toxic effects (e.g., substantial bone roethylnitrosoureas, such as streptozotocin (181,chlorozotocin marrow toxicity), they represent valuable drugs, since they (311,and MCNU (Cymerin, 34), which possess either a very low o r no bone marrow toxicity. However, it was are capable of entering the cerebrospinal fluid, and, hence, can be used for the treatment of brain tumors.9.15,26 reported3,'8,36.40.41,46,50,51 that some D-glucosamine derivaIn the search for more active and less toxic analogues of 1 tives exhibit diabetogenic properties in animals. Hence, the and 2, the replacement of the cyclohexyl moieties with the search for more active and less toxic carbohydrate derivatives five- and six-membered aminoxyl (nitroxyl) moieties led to continues. Over the years, a number of analogues exhibiting the discovery of 3 (SLCNU) and analogue 4 (see structure). It a wide scatter of activities have been synthesized. This type 0022-3549/9 1/0700-0693$01.OO/O 0 7991,American Pharmaceutical Association

Journal of Pharmaceutical Sciences I 693 Vol. SO, No. 7,July 7997

of data is difficult to interpret as a guide for a systematic approach to the design of new drugs. In this report, an attempt was made to rationalize these results on the basis of correlations of activities with corresponding lipophilicities. The detailed rationale for this approach has been delineated earlier.7J7925.33 Briefly, the approach is based on a design of anticancer drugs which permits a selective permeation of drugs through the cancerous cell membranes as opposed to the membranes of healthy cells.

compounds (18, 21, 24, 25, 28, 29, 31, and 34) are shown in Table 11. Biological Effects-Compounds 13-16, 18, 21, 24, 25, 28, 29, 31, and 34 were tested in vivo against the murine lymphocytic leukemia P388 in CD,F, male mice according to the National Cancer Institute Protocol.55Compounds 13, 14, and 34 at a dose of 20 mgkglday possessed outstanding activity, with %ILS values of 437,437, and 603, respectively. In the case of SLGNU (13) and analogue 14, all mice (6/6) were alive until the 60th day, and in the case of 34, all mice (616) were alive after 60 days. In contrast, the %ILS of the clinical Results and Discussion drug streptozotocin (18) and chlorozotocin (31),each at a dose In order to ascertain the position of the nitroso group in the of 8 mgkglday, were only 178 and 194, respectively, and all nitrosourea derivatives 13-16, a regioselective rneth0d23.25~54 mice succumbed within 30 days. was used to transfer the aminoxyl (nitroxyl) moiety containAs previously reported,25 the %ILS of the clinical drug 1 a t ing the nitroso group to the appropriate amine 11 or 12 to give a dose of 16 mgkglday was 182 and that for 2 was only 145 the corresponding nitrosoureas 13-16, respectively. Thus, 7 and all mice succumbed within 30 days. In contrast, 3 and and 8 were prepared by the condensation of the corresponding analogue 4, each a t a dose of 35 mg/kg/day, possessed spin-labeled isocyanates 5 and 6 with N-hydroxysuccinimide. outstanding activity with %ILS values of 542 and 514, Nitrosation of these active esters 7 and 8 with the dinitrogen respectively.25 In the case of 3, all mice (6/6) were alive after tetraoxide resulted in the formation of the corresponding 60 days, and in the case of 4, five mice (5/6) were alive after nitroso active esters 9 and 10. These spin-labeled nitroso 60 days. Even at a more toxic dose of 40 mgkglday, 13.14, and active esters 9 and 10 were reacted with either 11 or 12 to give 34 exhibited %ILS values of 230, 221, and 460, respectively the spin-labeled nitrosourea glucose derivatives 13-16 (Table 111). Compounds 15 and 16 had, respectively, %ILS (Scheme I) in 1 5 4 3 % yields. The yields and physical convalues of 174 and 64 a t a dose of 20 mglkgiday, and 156 and stants for 13-16 are presented in Table I. 41 at a 40 mg/kg/day dose (Scheme 111). All spin-labeled Attempts to prepare analogous spin-labeled nitrosoureas carbohydrate nitrosoureas 13-16 exhibited distinct activities derived from D-galactosyl-, D-mannosyl-, and D-lyxosylranging from low for 16 to a very high activity for 13 and 14, amines by the condensation of the amino sugars with the as evidenced by the %ILSvalues (Table 111). The %ILSvalues transfer agent 9 failed. Further, repeated attempts to synshown in Table 111also indicate that 24 and 28 (the galactosyl thesize the N-methyl-N-nitroso and the corresponding acyl and mannosyl analogues of streptozotocin, respectively) were derivative of D-lyxosylamine were also unsuccessful. marginally active, whereas their corresponding tetra-0The candidate pharmaceuticals 18, 21, 24, 25, 28, 29, 31, acetyl derivatives 25 and 29 were inactive. In addition, the and 343.35-37940,42.43950.53 derived from 2-amino-2-deoxy-D- glucamine derivative 21 was inactive. glucose (1 1), D-glucamine (19), 2-amino-2-deoxy-D-galactose Compounds 13-16,18,31, and 34 were then tested against (221, 2-amino-2-deoxy-D-mannose(261, a n d 6 the lymphoid leukemia L1210, and the results are shown in amino-6-deoxy-1-methoxy-D-glucose(32) were synthesized as Table IV. Compounds 13 and 14, a t the optimum dose of 50 delineated in Schemes I1 and 111, and detailed in the Expermglkglday, elicited %ILS values of 557 and 428, respectively, imental Section. The yields and physical constants for these whereas MCNU (34) elicited a %ILS of 713 a t the optimum 0 RNCO

HO-NSU

9 RN

RNH

L o - +

0

lo-+ R,o/!-

cCH1CI,/-35 H N :y:o

C

OR1

?!

0

l1

Or

l2

NO

I

NO

0

5, R = Re

7. R = R,

9. R = R,

6, R = RS

8 . R = R,

10, R = R5

HNTNR 0

13, R = Re. R1

= H

14. R = R,. RI

= H

15, R = Re. R1 = Ac 16, R = R,.

f OR’ HO-NSU =O-

b

‘.xJk I.

1 1 , R’ = H

R, =

12, R’ = Ac

0

Scheme I 694 I Journal of Pharmaceutical Sciences Vol. 80, No. 7, July 1991

O

R

1

R1 = Ac

Table I-Physical

Properties of Spin-Labeled Carbohydrate Nltrosoureas 13-16

Molecular Formulaa

Compound 13

Yield, %

mP (dec.),"C

18

oil

15

oil

43

174-1 79

32

172-1 76

C16H29N307

MS 345 (M+ 330 (M+ 331 (M' 316 (M+ 489 (M' 474 (M+ 475 (M' 469 (M'

(375.42) 14

C14H27N307

(361.39) 15

C22H37N3011

(519.54) 16

H,,N301

1

(505.52)

- 30,loo), - 45,62) - 30, loo),

- 45,58) - 30,loo),

- 45,42) - 30,loo), - 45,47)

a The microanalyses ((2, H, and N) were in satisfactory agreement with the calculated values within ?0.3%; MW is shown in parentheses. Mass spectra were obtained by using chemical ionization with methane as a reactant gas; relative percent intensities of the peaks are given.

NO

-

MeNCO RNH,

I

I

NaNQ

RNHKLIHMe ncl RNHKNHMa 0

0

18 l

l

HO +

-

kCH, H

l7

I

21

- -

25, R' = Ac

- -

29, R = k

Ac10

24, RI

H

Pyridine

R - H

26. R =

27. R

-

28. R

H

H

AolO

Pyn'dins

RO W - H

Scheme II

dose of 25 mglkglday and the clinical drug 18 had a %ILS of 55 a t a dose of 100 mglkglday.54 Chlorozotocin (31), at a dose of 30 mglkglday, exhibited a %ILS of 517. The previously reported25 %ILS value for the clinical drug 1, a t a dose of 25 mglkglday, was 646; the values for 3 and 4 at a dose of 60 mglkglday were 713 i2nd 612, respectively. In order to establish a measurable and predictable parameter for the correlation of the activities of the 13-16, 18, 21, 24, 25, 28, 29, 31, and 34 with the presumed permeation through cell membranes, a n attempt was made to relate the lipophilicities of these drugs to their structural features and to the levels of their anticancer activities. For this purpose, the partition coefficients for these compounds were determined in the n-octanol-water solvent system by using EPR and/or UV methodologies.56.57 The results are summarized in Table 111. The lower the P (or log P ) values, the higher is the hydrophilicity of a compound. Although all compounds studied (13-16,18,21,24,25,28,29,31, and 34) are water soluble, their partition coefficients varied over a wide range. There was good correlation, as expected, between the activity and lipophilicity of 13-16, 18, 24, 27, 28, 31, and 34 (Figure 1). Compounds 21, 25, ,and 29 were inactive and cannot be

NO

I

c, NNCO NoNO, RNHl

11. R =

NH-Ci

&OH

HCI

RNHy-C ),

0

0

30

31

33

34

HO

32, R =

&Me

HO

OH

Scheme Ill

meaningfully correlated with lipophilicities. According to the National Cancer Institute Protocol,55 in order for a compound Journal of Pharmaceutical Sciences I 695 Vol. SO,No. 7, July 7997

Table ICPhysical Properties of the Carbohydrate Nitrosoureas 18, 21, 24, 25, 28, 29, 31, and 34 Molecular Formula"

Compound 18

Yield, o/o

C8H15N307

MS

52

114-116

30

95-96

28

109-110

27

110-112

27

90-1 02

26

105-110

47

146-1 48

42

106112

(M+-106, 95) (M+-107, 100) (M+-121, 47) (M+-l06, 92) (M+-l07, 100) (M+-121, 33) (M+-l06, 94) (M+-107, 100) (M+-121,37) (M+-106, 82) (M+-107, 100) (M+-121, 35) (M+-l06, 93) (M+-l07, 100) (M+-121, 32) (M+-106, 82) (M+-107, 100) (M+-121, 28) (M+-81, 64) (M+-143, 55) (M+-242, 71) (MI-203, 68) (M+-204, 100) (M+-107, 51 (Mf-139, 86) (M+-219, 38) (M+-221, 100)

(265.22) 21

C6H17N307

(267.24) 24

C6H15N307

(265.22) 25

C16H23N,01

1

(433.37) 28

C8H15N307

(265.22) 29

C,6H23N301

1

(433.37) 31

C9H16CIN307 (313.69)

34

a The microanalyses (C, H, and N) were in satisfactory agreement with the calculated values within 50.3%; MW is shown in parentheses. Mass spectra were obtained by using chemical ionizations with methane as a reactant gas; relative percent intensities of the peaks are given.

Table ill-Partition Coefficients and Anticancer Actlvitles of 13-16, 18, 21, 24, 25, 28, 29, 31, and 34 against P388 Lymphocytlc Leukemia' In CDPF, Male Mice.

Compound

13 14 15 16 18 21 24 25 28 29 31 34

Dose, mglkglday

20 40 20 40 20 40 20 40 8 16 8 16 16 8 16 16 8 10 20 40

TIC. '10

537 330 537 321 274 164 256 141 278 110 112 141 109 111 133 110 294 551 703 560

%ILS

437 230 437 221 174 64 156 41 178e 10 12 41 9 11 33 10 1949 451 603 460

Partition Coefficientd

5-Day Weight Change, Yob

Cures (Survivors1 Totals)c

+4.6 -5.1 +4.4 -5.8 +9.8 -2.3 +8.4 -2.8 -8.8 +5.8 +2.2 +3.8 f5.5 +2.4 +3.6 +2.6 -5.4 +3.8 -4.2 -5.1

616 016 616 016 016 0/6 016 016 016 016 016 016 016 016 016 0/6 016 516 616 516

Log P

P

ESR

UV

ESR

uv

79

74

1.89

1.87

66

64

1.82

1.81

189

186

2.28

2.27

180

178

2.56

2.25

-

0.14 0.1 1

-0.85' -0.96

-

0.13 - 724

-0.89 2.86

-

-0.92 2.91 - 1.52'

0.12 - 813 0.03

-

17.4

1.24

a P388 tumor; NationalCancer InstituteProtocol (ref 55) followed exactly. The average percentage weight change on day 5 was taken as a measure of drug toxicity. Cures mean survival after 60 days. dThe partition coefficients (P = [compound in 1-octanol]/[compoundin water]) were obtained by ESR and UV techniques, according to literaturemethods (refs 56 and 57). In ref 60, the %ILS is 185. In ref 61, the log Pvalues for streptozotocin (18) is -0.89 and that for chlorozotocin (31) is -1 55. 9 In ref 60, the %ILS is 210.

'

to be considered active, it must have a %ILSvalue of at least 25. Nonetheless, the inactivity of 25 and 29 can be attributed to their relatively high hydrophobicity, whereas that of 2 1 can be attributed to its relatively high hydrophilicity (Table 111). Compounds 24 and 28 were marginally active, correlating with high hydrophilicity. However, the hydrophilic clinical 696 I Journal of Pharmaceutical Sciences Vol. 80, No. 7, July 1991

drugs 18 and 31 were more active than 24 and 28 (Table 111). This discrepancy might be attributed to difficulties in correlating marginally active compounds with lipophilicities. Compound 34, with the highest activity of all tested compounds, exhibited slight hydrophobicity. Compounds 3 and 4 were more hydrophobic and somewhat less active than 34.

Table IV-Anticancer Actlvltles of 1S16, 18, 31, and 34 Agalnst L1210 Lymphold Leukernla' In CD,F, Male Mice

Compound

Dose, mglkglday

T/C,

%ILS

5-Day Weight Change, %b

Cures (Survivors1 Totals)'

13

6.3 12.5 25 50 100

103 124 174 657 123 528 109 280 142 27 1 129 155 617 290 274 813

3 24 74 557 23 428 9 180 42 171 29 55d 517e 190" 174' 7139

+3.4 +4.8 +6.2 -9.1 -2.2 -7.8 -1.6 +6.2 -2.8 -5.6 +3.2 +2.6 -4.4 -1.6 -2.2 -3.2

016 016 016 616 016 516 016 016 016 016 016 016 616 316 416 616

14

50

15

100 20

16

20

18 31 34

100 30 6.3 12.5 25

Y O

50 50

a L1210 tumor; National Cancer Institute Protocol (ref 55) followed exactly. The average percent weight change on day 5 was taken as a measure of drug toxicity. Cures mean survival after 60 days. In ref 60, the %ILS is 62. " In ref 48, the %ILS is 184. ' In ref 48, the Y'ILS is 167. 9 In ref 48, comdete cure is observed.

Compounds 13 and 14 were less active than 3 and 4 and, as expected, correspondingly more hydrophobic (Figure 1).The acetylated derivatives 15 and 16 and the clinical drugs 1 and 2 were more hydrophobic than 13 and 14 and less active (Figure 1).Thus, from this correlation, it is evident that neither extreme hydrophilic nor hydrophobic properties correlate with optimum activity and that the most active drugs will be found between these extremes, as in the case of 34 and 3. In conclusion,it was shown that scattered literature results of various levels of activity of nitroso and chloroethylnitroso derivatives of carbohydratescan now be explained by a linear correlation of lipophilicities with antileukemic activities (Figure 1). Furthermore, predictions can be made for future systematic studies of analogous carbohydrate derivatives to improve the parameters of toxicity and activity.

Experimental Section Materials-All reagents were of the best quality commercially available. Solvents were dried by standard procedures.58 Analytical Procedures-All melting points were obtained with a Thomas Hoover capillary melting point apparatus (model 6406-K) using a calibrated thermometer. Mass spectra were recorded on a 1000 1

t i

A

1 u

HYDROPHILIC

18

loo\

- t+-b- - - 'Ol

i -2.0

HYDROPHOBIC

I

\

I i

-1.0

0.0

1 .o

i

:

:

I

i

2.0

:

;

:

I

3.0

~

i

'

'

4.0

P Figure 1-Anticancer activity-lipophilicity correlations of urea analogues hydrophobic. The dotted line in of nitrosoureas. Key: (W) hydrophilic;(0) the hydrophilic range is an averaging linear regression line (ref 33) which is used here only to indicate a general trend. LOG

Hewlett-Packard mass spectrometer (model 5985GS) using a direct insertion probe, a source pressure of 2 x torr, and methane as a reactant gas for chemical ionization. Microanalyses were performed on a PerkingElmer 240 C elemental analyzer. Column chromatography was performed using the flash column chromatography techn i q ~ on e ~ silica ~ gel 60 (Fluka) finer than 230 mesh. The TLC analyses were performed on silica gel 60FZ5, precoated sheets (EM reagents; layer thickness 0.2 mm), with visualization using UV light and/or iodine chamber. The purity of 13-16 was checked in a solvent system composed of methylene chloride and methanol (9:1, vlv). The analytical data for 13-16 are presented in Table I. For measuring the partition coefficients, 1-octanol and water layers were presaturated with each other prior to use. Partition coefficients (P)were obtained by following literature methodologyF6.57 using EPR and/or UV spectrophotometry. Thus, the areas of the octanol solutions and the separated water solutions were used to compute the concentrations of 13-16, 18, 21, 24, 25, 28, 29, 31, and 34 in the octanol and water layers. The partition coefficients (P = compound in 1-octanoll compound in water) and their logarithmic values so obtained are shown in Table 111. Mice-Male mice (CD,F, for testing, average weight 18-20 g1mouse; and DBN2 for tumor propagation,55 6-7 weeks old) were supplied by Harlan Sprague-Dawley, Indianapolis, IN. The tumor lines P388 and L1210 cells were supplied by DCT Tumor Repository, NCI Frederick Cancer Research Facility, Frederick, MD. Mice were fed Rodent Laboratory Chow 5001 (Ralston Purina Company) and water ad libitum. Biological Evaluations-Compounds 13-16,18,21,24,25,28,29, 31, and 34 were evaluated in vivo against the lympholytic leukemia P388 and 13-16, 18, 31, and 34 were tested against lymphoid leukemia L1210 in mice following the protocol of the National Cancer Institute.55 The CD,F, male mice of 18-20 g weight, in groups of six, were inoculated ip with either lo6 cells of P388 tumor or with lo5 cells of L1210 tumor on day zero of the experiment. Compounds 13-16,18, 21,24,25,28,29,31, and 34 for P388 and 13-16,18, 31, and 34 for L1210 evaluations were injected ip a t doses listed in Tables I11 and IV, respectively, for 9 successive days starting from day one. The animals were then observed according to a previously published protocol56 for 30 or 60 days, keeping a record of deaths and survivors. The anticancer activity was evaluated by comparing the mean survival time of the treated mice with that of the control animals be., by the TIC method, where T represents the mean survival time of the treated group and C the mean survival time of the tumor-bearing control group). The %ILS parameter was calculated by the formula [(T--C)/ Cl x 100. The results of the %TIC and %ILS values are summarized in Tables I11 and IV for P388 and for L1210 mice, respectively. Preparation of N'-Hydroxysuccinimide-N-substitutedCarbamates 7 and 8-A General Procedure-To a stirred solution N-hydroxysuccinimide (1.15 g, 10.00 mmol) in acetonitrile (20 mL), either 5 Journal of Pharmaceutical Sciences I 697 Vol. 80, No. 7, July 7997

(1.97 g, 10.00 mmol) or 6 (1.83 g, 10.00 mmol) in acetonitrile (15 mL) y1)urea (31)-A General Procedure-To a solution of either 17 (1.74 was added in a dropwise manner over a period of 15-20 rnin at 0 "C. g, 7.38 mmol) or 30 (1.82 g, 7.38 mmol) and sodium nitrite (1.00 g, After the addition, the reaction mixture was stirred for 4 h at 0 "C and 15.76 mmol) in aqueous ethanol (60 mL, water:ethanol = 2:1, vlv) was then for 16 h a t 25 "C. The reaction mixture was concentrated on a added concentrated HCl(4mL) in a dropwise manner over a period of rotating evaporator at 40 "C/20 torr to afford either 1.25 g (40%) of 7 10 to 15 min at 0 "C. The solution was stirred for 30 rnin at 0 "C and as a n oily, red liquid or 1.13 g (38%) of 8 as an oily, yellow liquid. then for 1h at 25 "C. The resultant clear yellow solution was cooled Preparation of N'-Hydroxysuccinimide-N-substituted Nitrosoto 0 "C and the deposited ivory colored crystals were filtered under carbamates 9 and 10-A General Procedure-To a stirred solution of reduced pressure a t 25 "C. The crystals were washed with anhydrous either 7 (1.25 g, 4.0 mmol) or 8 (1.19 g, 4.0 mmol) in dry THF (10 mL) ether (2 x 5 mL). Thus, either 18 or 31 was obtained. The yields and was added anhydrous sodium acetate (1.97 g, 24.4 mmol). The analytical data are presented in Table 11. mixture was cooled to -34 "C. To this cooled and stirred mixture, a Pre pa ra tion of l-Methyl-3-(l-deoxy-~-glucamin-l-yl)urea solution of dinitrogen tetraoxide (0.4 g, 4.36 mmol) in dry carbon ( 2 0 h T o a solution of 19 (2.25 g, 12.4 mmol) in 1M NaOH (17.5 mL) was added a solution of methylisocyanate (0.71 g, 12.4 mmol) in a tetrachloride (10 mL) was added in a dropwise manner over a period of 15-20 min. After the addition, the reaction mixture was stirred for mixture of chloroform and ethyl ether (150 mL, 1:1, v/v) at 0 "C. The 1h a t -35 "C and then for 30 min at 5 "C. To the reactionmixture was reaction mixture was stirred vigorously at 0 "C for 3 h and then at 25 "C for 3 h. The resulting precipitate was filtered and washed with then added a mixture of methylene chloride (10 mL) and ice water (6 anhydrous ether (2 x 15 mL). Repeated recrystallization of the solid mL). The organic layer was separated and the aqueous layer was extracted with methylene chloride (4 x 10 mL). The combined organic from absolute ethanol gave 1.92 g (65%) of pure 20 as pale yellow crystalline material; mp (dec.) 126-127 "C (lit. mp 125-127 "C40). extracts were washed with 5% aqueous sodium bicarbonate solution (2 x 5 mL) and then with water (2 x 5 mL). The organic solution was Preparation of l-Methyl-l-nitroso-3-(l-deoxy-D-glucamin-ldried over anhydrous magnesium sulfate and filtered. Concentration y1)urea ( 2 1 b T o a solution of 20 (1.76 g, 7.38 mmol) and sodium of the filtrate on a rotating evaporator at 40 "C/20 torr gave the crude nitrite (1.00 g, 15.76 mmol) in aqueous ethanol (60 mL, water:ethanol products. Purification by flash column chromatography on silica gel = 2:1, v/v) was added concentrated HCl(4 mL) in a dropwise manner using methylene chloride and methanol (95:4, v/v) as eluant and over a period of 10 to 15 min at 0 "C. The solution was stirred for 30 min a t 0 "C and then for 1 h at 25 "C. The resultant clear yellow subsequent concentration of the combined fractions on a rotating evaporator at 24"C/20 torr gave either 0.50 g (37%) of the pure solution was cooled to 0 "C and the deposited pale yellow crystalline material was filtered and washed with anhydrous ether (2 x 5 mL) product 9 or 0.44 g (34%) of the pure product 10 as oily liquids. Preparation of 1-(2,2,6,6-Tetramethylpiperidin-l-oxy-4-yl)-3-(2- to give pure 21. The yield and analytical data are presented in Table deoxy-~-glucos-2-yl)-l-nitrosourea (13), and 1-(2,2,5,5-Tetrameth11. ylpyrrolidin- 1-oxyI-3-yl)-3-(2-deoxy-D-glucos-2-yl)-l-n~trosourea Preparation of l-Methyl-3-(2-deoxy-D-galaetos-2-yl)urea(23) ( 1 4 b A General Procedure-A solution of either 9 (0.48 g, 1.40 mmol) and l-Methyl-3-(2-deoxy-D-mannos-2-yl)urea ( 2 7 h A General Procedure-To a solution of either 22 (2.67 g, 12.4 mmol) or 26 (2.675 g, or 10 (0.46 g, 1.40 mmol) and 11 (0.24 g, 1.35 mmol) in methylene chloride (20 mL) was stirred for 2 h at 5 "C and then for 15 h at 25 "C. 12.4 mmol) in 1M NaOH (17.5 mL) was added methylisocyanate (0.71 g, 12.4 mmol) in a mixture of chloroform and ethyl ether (150 mL, 1:1, The reaction mixture was washed with a saturated NaCl solution (2 x 15mL), dried over anhydrous magnesium sulfate, and filtered. The v/v) a t 0 "C. The reaction mixture was vigorously stirred for 3 hat 0 "C filtrate was concentrated on a rotating evaporator a t 25 "C/20 torr. and then for a n additional 3 h at 25 "C. The resultant precipitate was The resulting residue was purified by flash chromatography on silica filtered and the solid successively washed with chloroform (2 x 10 mL), ethyl ether (2 x 10 mL), and petroleum ether (92 x 10 mL). gel using methylene chloride and methanol (95:5, v/v) as eluant. Concentration of the combined fractions on a rotating evaporator a t Thus, either 23 (1.82 g, 77% yield) or 27 (1.83 g, 77% yield) was obtained. 25 "C/20torr gave either 0.10 g (18%)ofpure 13or 0.08 g (15%)ofpure 14 as viscous oils. The yields and analytical data are presented in Preparation of l-Methyl-l-nitroso-3-(2-deoxy-D-galactos-2y1)urea (24) and l-Methyl-l-nitroso-3-(2-deoxy-D-mannos-2-yl)urea Table I. Preparation of 1-(2,2,6,6-Tetramethylpiperidin-l-oxyl-4-y1)-3-(2- ( 2 8 b A General Procedure-To a solution of either 23 (1.74 g, 7.38 deoxy-l,2,4,6-tetra-O-acetyI-D-glucos-2-yl)-l-n~trosourea (15) and mmol) or 27 (1.74 g, 7.38 mmol) and sodium nitrite (1.00 g, 15.76 1-(2,2,5,5-Tetramethylpyrrolidin-l-oxyl-3-yl)-3-(2-deoxy-1,3,4,6- mmol) in aqueous ethanol (60 mL, water:ethanol = 2:1, v/v) was tetra-0-acetyl-~-glucos-2-yl)-l-nitrosourea (l6)-A General Proceadded concentrated HCl(5 mL) in a dropwise manner over a period dure-A solution of either 9 (0.48 g, 1.40 mmol) or 10 (0.46 g, 1.40 of 10-15 rnin a t 0 "C. The solution was stirred for 30 rnin at 0 "C and mmol) and 12 (0.47 g, 1.35 mmol) in methylene chloride (20 mL) was then for 1h a t 25 "C. The resultant clear yellow solution was cooled stirred for 2 h a t 50 "C and then for 15 h at 25 "C. The work-up to 0 "C. A pale yellow, semisolid precipitate was obtained which was procedure was identical to that described for 13 and 14. The yields and collected by filtration and purified by flash chromatography on silica gel using chloroform and methanol (3:1, v/v) as eluant. Concentration analytical data are presented in Table I. Attempted Preparation of 1-(2,2,6,6-Tetramethylpiperidin-l- of the combined fractions containing the product on a rotating oxy~-4-y~)-3-(2-deoxy-D-galactos-~-y~)-~-nitrosourea, 1-(2,2,6,6evaporator a t 25 "C/20 torr gave either pure 24 or 28. The yields and Tetramethylpiperidin-1-oxyl-4-yl)-3-(2-deoxy-D-mannos-2-y1)-1- analytical data are presented in Table 11. nitrosourea, and 1-(2,2,6,6-Tetramethyl-piperidin-l-oxy1-4-y1)-3- Preparation of l-Methyl-l-nitroso-3-(1,3,4,6-tetra-0-acetyl-2(I-deoxy-D-lyxos-1-y1)-I-nitrosourea.-A General Procedure-A deoxy-D-galactos-2-y1)urea(25) and l-Methyl-l-nitroso-3-(1,3,4,6solution of 9 (0.48 g, 1.40 mmol) and the corresponding amino sugar tetra-O-acetyl-2-deoxy-D-mannos-2-yl)urea (29)-A General Proce(1.35 mmol) in methylene chloride (20 mL) was stirred for 2 h at 5 "C dure-A suspension of either 24 (1.326 g, 5 mmol) or 28 (1.326 g, 5 and then for 15 h at 25 "C. The work-up procedure was identical to mmol) in a mixture of acetic anhydride (15 mL, 0.16 mol) and dry pyridine (27 mL) was stirred for 2 h at 0 "C. The resultant pale yellow that described for 13 and 14. However, attempts to obtain the pure solution was poured into cold water (150 ml) and the aqueous solution target compounds were not successful. Preparation of l-Methyl-3-(2-deoxy-D-glucos-2-yl)urea (17) and was cooled for 1h in a n ice bath mixture at 0-5 "C.The precipitated 1-(2-Chloroethyl)-3-(2-deoxy-D-glucose-2-yI)urea ( 3 0 b A General solid was filtered and washed with ethyl ether (3 x 5 mL). Repeated Procedure-To a solution of 11(2.675 g, 12.4 mmol) in 1M NaOH (17.5 recrystallization from ethanol gave either 25 or 29 as pure products. mL) was added either a solution of methylisocyanate (0.71 g, 12.4 The yields and analytical data are presented in Table 11. mmol) or a solution of 2-chloroethylisocyanate (1.31 g, 12.4 mmol) in Preparation of l-(2-Chloroethyl)-3-(l-methyl-6-deoxy-cu-Da mixture ofchloroform and ethyl ether (150 mL, 1:1, v/v) at 0 "C. The glucopyranos-6-yl)urea ( 3 3 b T o a solution of 32 (2.40 g, 12.4 mmol) heterogeneous suspension was stirred vigorously a t 0°C for 3 h. in 1M NaOH and sodium nitrite (1.00 g, 15.76 mmol) in aqueous During this period, a white precipitate was formed. The precipitate ethanol (60 mL, water:ethanol = 2:1, v/v) was added concentrated was filtered under reduced pressure at 25 "C and washed successively HCl(4 mL) in a dropwise manner over a period of 10 to 15min at 0 "C. with chloroform (2 x 15 mL), ethyl ether (2 x 15 mL), and petroleum The reaction mixture was stirred vigorously for 4 h at 0 "C and then ether (2 x 15 mL). Thus, either 17 L1.84 g (78%)mp (dec.) 124-126 "C] for an additional 4 h at room temperature (25 " 0 . The resultant pale yellow crystalline material was filtered and washed with anhydrous or 30 12.135 g (75%), mp (dec.) 148-150 "C (lit. mp (dec.) 150 "C42 or 160-161 "C31was obtained. ether (3 x 5 mL) to give 2.22 g (60%) of pure 33 as a pale yellow P r e p a r a t i o n of l-MethyI-l-nitroso-3-(2-deoxy-D-glueos-2crystalline material; mp (dec.) 142-144 "C. yl)urea (18) and l-(2-Chloroethyl)-l-nitroso-3-(2-deoxy-D-glucos-2- Preparation of l-(2-Choroethyl)-l-nitroso-3-(l-methyl-6-deoxy698 I Journal of Pharmaceutical Sciences Vol. 80, No. 7, July 1997

cr-D-glucopyranos-6-y1)urea( 3 4 b T o a solution of 33 (2.22 g, 7.38 mmol) and sodium nitrite (1.00 g, 15.76 mmol) in aqueous ethanol (60 mL, water:ethanol = 2 I, vlv) was added concentrated HCl(4 mL) in a dropwise manner over a period of 10 to 15 min at 0 "C. The solution was stirred for30 min at 0 "C and then for 1.5h a t 25 "C. The resultant pale yellow compound was filtered and washed with anhydrous ether (3 x 5 mL) to give the pure 34. The yield and analytical data are presented in Table 11.

References and Notes 1. Johnson, T. P.; McCaleb, G. S.; Opliger, P. S.; Montgomery, J. A. J . Med. Chem. 1966,9, 892. 2. Johnston, T. P.; McCaleb, G. S.; Opliger, P. S.; Laster, W. R.; Montgomery, J . A. J. Med. Chem. 1971,14,600. 3. Johnston, T. P.; McCaleb, G. S.; Montgomery, J. A. J . Med. Chem. 1975,18,104. 4. Mongtomery, J . A.; Thomas, H. J . J . Med. Chem. 1979,22,1109. 5. Montgomery, J. A.; Mayo, J. G.; Hansch, C. J.Med. Chem. 1974, 17, 477. 6. Crider, A. M.; Lamey, R.; Floss, H. G.; Cassady, J. M.; Bradner, W. J. J. Med. Chem. 1980,23, 848. 7. Hansch, C.; Leo, A,; Schmidt, C.; Jow, P. Y. C.; Montgomery, J . A. J . Med. Chem. 1980,23, 1095. 8. Schmahl, D. Maligne Tumoren; Editio Cantor Aulendorf, 1981. 9. Medical Oncology; Calabresi, P.; Schein, P. S.; Rosenberg, S. A., Eds.; Macmillan: New York, 1985. 10. Lown, J. W.; Chen, H. H. Can J . Chem 1981, 59, 390, and references therein. 11. Umezawa, H. In Medicinal Chemistry; Cassady, J. M.; Douros, J., Eds.; Academic: New York, 1980; pp 146-148, and references therein. 12. Montgomery, J. A.; James, R.; McCaleb, G. S.; Kirk, M. C.; Johnston, T. P. J. Med. Chem. 1975,18, 568. 13. Lown, J. W.; Chauhan, S. M. S. J . Org. Chem. 1981,46,2479,and references therein. 14. Montgomery, J. A. In Nitrosoureas in Cancer Treatment; Serrou, B.; Schein, P. S.; Imbach, J. L., Eds.; Elsevier North-Holland Biomedical: Amsterdam, 1981; p 13, and references therein. 15. Wasserman, T. H.; Slanik, M.; Carter, S. K. Cancer 1975, 36, 1258. 16. Wheeler, G. P.; Johnston, T. P.; Bowdon, B. J.; McCaleb, G. S.; Hill, D. L.; Montgomery, J. A. Biochem. Pharmacol. 1977, 26, 2331. 17. Hansch, C.; Smith, N.; Engle, R.; Wood, H. Cancer Chemother. Rev. 1972. 56. 443. 18. Schein, P.'S.; McMenamin, M. G.; Anderson, T. CancerRes. 1973, 33, 2005. 19. Wheeler, G. P.; Bowdon, B. J.; Grimsley, J. A.; Lloyd, H. H. Cancer Res. 1974, 34, 194. 20. Panasci, L. C.; Green, D.; Nagourney, R.; Fox, P.; Schein, P. S. Cancer Res. 1977,37, 2615. 21. Eisenbrand, G.; Miiller, N.; Denkel, E.; Sterzel, W. J . CancerRes. Clin. Oncol. 1986, 112, 196. 22. Raikov, Z.; Todorov, D.; Ilarianova, M.; Denisov, G.; Tsanova, T.; Kafalieva, D. Cancer Biochem. Biophys. 1985, 7, 343. 23. Sosnovsky, G.; Li, S. W. Drugs of the Future 1985,10,211. 24. Sosnovsky, G.; Li, S. W. Life Sci. 1985,36, 1479. 25. Sosnovsky, G.; Li, S. W.; Rao, N. U. M. Z. Naturforsch. 1987,42c, 921. 26. Carter, S. K.; Bakowski, M. T.; Hellmann, K. Chemotherapy of Cancer, 3rd Ed., Churchill Livingston: New York, 1987. 27. Suami, T.; Kato, T.:, Hisamatsu, T. In Nitrosoureas in Cancer Treatment; Serrou, B.; Schein, P. S.;Imbach, J. L., Eds.: Elsevier North-Holland Biomedical: Amsterdam, 1981, and references therein. _ . ... . . .

28. Gutierrez, P. L.; Konieczny, M.; Sosnovsky, G. Z. Naturforsch. 1981,366, 1612. 29. Afzal, V.; Brasch, R. C.; Nitecki, D. E.; Wolq S. Invest. Radiol.

1984, 19, 549. 30. Sosnovsky, G.; Li, S. W. Cancer Lett. 1985,25, 255. 31. Ross, D. D.; Gutierrez, P. L.; Sosnovsky, G. Am. Assoc. Cancer Res. Preclinical Pharmncol. Exp. Ther. 1984, Abstract 1158. 32. Brasch, R. C.; Nitecki, D. E.; Brant-Zawadzki, M.; Enzmann, D. R.; Wesbey, G. E.; Tozer, T. N.; Tuck, L. D.; Cann, C. E.; Fike, J. R.; Sheldon, P. Am. J . Neurol. Radiol. 1983,4, 1035. 33. Sosnovsky, G. Pure Appl. Chem. 1990, 62, 289, and references therein. 34. Ogawa, M. In Nitrosoureas; F'restayko, A. W.; Baker, L. H.; Crooke, S. T.; Carter, S. K.; Schein, P. S.; Alder, N. A., Eds.; Academic: New York, 1980; p. 399, and references therein. 35. Herr, R. R.; Jahnke, H. K.; Argoudelis, A. D. J.Am. Chem. SOC. 1967,89,4808. 36. Hardegger, E.; Meier, A.; Stoos, A. Helv. Chim. Acta 1969,52, 2555. 37. Hessler, E. J.; Jahnke, H. K. J . Org. Chem. 1970,35, 245. 38. Machinami, T.; Suami, T. Bull. Chem. SOC.Jpn. 1973,46, 1012. 39. Suami, T.; Machinami, T. Bull. Chem. SOC.Jpn. 1970,43,3013. 40. Bannister. B. J . Antibiot. 1972.25. 377. 41. Meier, A.;~Stoos,F.; Martin, D.: Buyuk, G.; Hardegger, E. Helv. Chim. Acta 1974,57,2622. 42. Burns, H. D.; Heindd, N. D. Org. Preps. Proc. Zntl. 1974, 6, 259. 43. Gassmann, N.; Stoos, F.; Meier, A.; Buyuk, G.; Helali, S. E.; Hardegger, E. Helv. Chim. Acta 1975,58, 182. 44. Machinami, T.; Kobayashi, K.; Hayakawa, Y.; Suami, T. Bull. Chem. Soc. Jpn. 1975,48,3761. 45. Machinami, T.; Nishiyama, S.; Kikuchi, K.; Suami, T. Bull. Chem. Soc. Jpn. 1975,48,3763. 46. Fujiwara, A. N.; Acton, E. M.; Henry, D. W. J . Med. Chem. 1976,

."

mn'7

1 I , GYZ.

47. Hisamatsu, T.; Uchida, S. Gann 1977, 68, 819. 48. Sekido, S.; Ninomiya, K.; Iwasaki, M. Cancer Treat. Rep. 1979, 63,961. 49. Bernacki, R. J.; Sharwa, M.; Angelino, N.; Korytnyk, W. Proc. Am. Assoc. Cancer Res. 1982,23, 204. 50. Bernacki, R. J.; Wilson, G. L.; Mossman, B. T.; Angelino, N.; Kauter, P. M.; Korytnyk, W. Cancer Res. 1985,45, 695. 51. Roger, P.; Monneret, C.; Fournier, J.-P.; Choay, P.; Gagnet, R.; Gosse, C.; Letournew, Y .; Ahassi, G.; Gouyette, A. J . Med. Chem. 1989,32, 16. 52. Hoshi, W. Drugs of the Future 1986,11, 261. 53. Hoshi, A. Drugs of Today 1988,24, 87, and references therein. 54. Sosnovsky, G.; Rao, N. U. M. Drugs of the Future 1986,11,285. 55. Geran, R.; Greenberg, N.; Macdonald, M.; Schumacher, A.; Abbott, B. Cancer Chemother. Rep. Part 3 1972,3, 1-103. 56. Keana, J. F.; Lee, T. D.; Bernard, E. M. J . Am. Chem. Soc. 1976, 98, 3052. 57. Purcell, W. P.; Bass, G. E.; Clayton, J. M. Strate y of Drug Design: A Guide to Biological Activity;John Wiley an! Sons: New York, 1973, pp 131-133. 58. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of Laboratory Chemicals; Pergamon: New York, 1966. 59. Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978,43, 2923. 60. National Cancer Znstitute Monogra h 55, Ex erimental Evaluation of Antitumor Drugs in the &A and 6 S S R and Clinical Correlations; National Cancer Institute: Bethesda, MD, 1980. 61. Hansch, C.; Leo, A. Substituent Constants for Correlation Analysis in Chemistry and Biology; John Wiley and Sons: New York, 1979: pp 246-258.

Acknowledgments The authors thank the Elsa U. Pardee Foundation, Midland, MI, for generous financial support of this work. One of the authors (N.U.M.R.) is grateful for partial sup ort by the discretiona funding of G.S. The authors are also indebteato the Tokyo Tanabezompan Ltd., Tokyo, J a .an, for Sam les of MCNU (C merin) and to HUT; for a sample ofYD-glucamine. America, Inc., Ascataway ,

&,

Journal of Pharmaceutical Sciences I699 Vol. SO,No. 7, July 7991