ClI.\lW
I
",
2"
"9 A
O
" V F O N
A
N
I
I
TroH2c@ OH 5
H, tri 6
J. J
which contained the calculated amount of phosphate upon digestion u-ith 5'-nucleotidase. The two phosphates were interrelated by conversion of 2 to 4 with nitrous acid.20 An alternative chemical synthesis was carried out by conventional methods.21 Tritylation of 3 at elevated temperaturez2with an excess of triphenylmethyl chloride gave a mixture of oily 2'!.5'- and 3',5'-ditrityl derivatives (6), with the latter predominating. These derivatives \vere separated by chromatography on silica gel, acetylated, and detritylated to the 3'- and 2'-monoacetates (7), respectively, the latter emerging crystalline. A room-temperat ure tritylation, hoLvever, gave the desired crystalline monotrityl derivative ( 5 ) , characterized as the 5' isomer by nmr. This in turn permitted selective acetylation of the secondary hydroxyl groups to give the crystalline diacetate (9) after detritglation of the iriterniediate trityl diacetate 8 (not isolated). Phosphorylation by the method of T e ~ i e rgave ~ ~ an intermediate cyanoethyl phosphate (10) which \vas treated with sodium methoxide in methanol24to remove protecting groups, again giving the desired .5'-phosphate (4) identical with the product from enzymatic transphosphorylation. The over-all yield from these several steps was loyo. Although the enzymatic synthesis undoubtedly gives higher yields than the chemical multistep sequence, the latter is preferred if larger quantities of material are desired. This conclusion rests mainly on the fact that it is difficult to separate the products from the surviving phosphate donor. Severtheless, it is hoped that technical advances will make such simple one-step enzyniological conversion practical for the organic chemist. Biological Activity.-The nucleotide analogs here discussed were tested in viuo against the transplanted Sarcoma 180 tumor in the mouse.25 As seen in Table I, TABLEI: AcuvI,rY .E.\INST
SAI~WMA 180 IN &Irma Index, C / T
Compound
ANI
O
HoH2cQJ
- '""'GS1 0 H, Ac
Ac0
8
1
\
0
H N y F O N
AI
c
ara-FC 16.7 am-FC 5'-phosphate 4.6 ara-FU 3.0 ara-FU j'-phosphate 3.7 (L Dosage, 50 m g of iiricleoside or it:, molar equivaleiit of phosphate/kg ip. The index C / T deiioter ratio of tumor weights, cont r d 5 / t reat ed aiiimals.23 (20) .iIileinaeller, . Biockem. J . , 36, i 2 9 (1942). ('21) .\. 11. Michelson, "The Chemistry of Nucleosides and Nucleotides," .\cademic Press Inc., Piev York, K. Y . , 1963, p 126 ff. (22) After J. P. Horvitz, J. .i. Urbanski, and J. Chua, J . O r y . Ciiem., 27, 3300 (1962). (23) G . 11. Tener, J . A m . Chem. Soc., 83, 165 (1961). (2.1) These conditions, rather t h a n use of t h e conventional aqueous alkali,
were chosen in order t o prevent t h e well-knoivn sensitivity of such arabinosides toward ring opening [J. J. Fox, N. C. Miller, and R. J. Cusliiey, Tetrahedron Letter.?, 4927 (1966)l. N o such opening v a s observed when a model (I~-~-arabinosy1-5-fluorouracil tribenzoate) was sul)jected to methanolysis, presumabl? because t h e postulated intermediate of this potentially destructil-e pathway is not prevented from reclosure ( a ) , in contrast t o t h e hydrolysis groduct (h), xhicli carries a negative charge.
10
-
-0
OCHB
AcO
I
I
9
Tr = C(C,,H,), AC= COCHj a
free acid v i t h Dowex-50 (H+) loivered the yield to 20% of paper chromatographically lioniogeiieous material
b
( 2 5 ) Procedure as in E. Grunberg. €1. N . Prince, E. Titsn'orth, G . lieskid, and 11. D. Tendler, Chemotherapia. 11, 249 (1966). \\ e wish tu thank L)r. Grunljerg a n d his staff fur making these rezultb avaiIai,Ie tu us.
I. 2. 11.
c. .I.
ti. I .
pho~phorylatioiidoes riot abolish the activity observtttl foi. the precursor iiucleositle analogs. However, xvhile t I i c w appears to be little effect upoii phosphorylatioii of ura-I'U, the more interestiiig am-FC seems to hecoiiie less active as t h e iiucleotide, at least at t h e dosage st ritlicd.
May 1968
~@-D-ARABIXOSYL-.~-FLUOHOPYRIMIDINE KUCLEOTIDES
nucleoside analog), nmr spectrum, and uv spectrum (100e~s= 8600, pH 7). Interconversion.-The hydrolytic deamination followed the procedure of Kleinzeller.20 The 5'-phosphate of ara-FC (2, t5 @mol)was dissolved in 64 p1 of Hz0. The following was added, in the order specified: KaOAc ( 5 . 5 mg), AcOH (20 pl), HzO (40 pl), arid ?;aNOI (60 mg). The resulting solution (pH 3.7) was allowed t,o staiid at, room temperature. After various time intervals, 12.5-pI aliqriots were spotted on What,man No. 3 filter paper and siibjected to descending paper chromatography in system B. After 1 hr, S.iC.'O of the st,arting material 2 had been convei ted to the analogoiis fluorouracil derivative 4 as indicated by migration against standards. Chemical Synthesis. 5'-Triphenylmethyl-lp-~-arabinofuranosyl-5-fluorouracil (5).--ara-FU (3, 2.72 g, 10.4 mmol) was dissolved in 31 ml of anhydrous pyridine, triphenylmethyl chloride (3.62 g, 13.0 mmol) was added, and the mixture was stirred at' room temperature for 65 hr. At the end of that period an equal volrime of ice water was added, and the mixture was concentrated to a syrup in uacuo. The residue was partitioned between 100 ml of H 2 0 and an equal volume of CHCI,. The organic layer was dried (NaZf-304) and Concentrated to a small volume to give 3.39 C, H, ?i. g (6.73 mmol) of 5, mp 124-126'. Anal. (Cz8HnjFX2OS) Absorption peaks of spectra (uv, ir, nmr) were as expect,ed. Ditritylation of lp-~-Arabinofuranosyl-5-fluorouracil.-Triphenylmethyl chloride (16.4 g, 59 mmol) was dissolved in 25 ml of anhydrous pyridine and am-FU (3, 1.28 g 4.9 mmol) was added to the solution previously heated to 95'. Heating was continued for 1 hr, resulting in slight browning of the initially clear yellow solution. Pyridine was removed in uaciio and the residiie was partitioned between water (2.5 ml) and CHCI, ( 5 0 ml). The aqueous port,ion was back-extracted with CHC1, ( 2 5 ml each portion), atid the combined organic layers m-ere concentrated to a dry residue. Chromatography on silica gel gave, after removal of reagent with benzene, two oily materials which were eluted with CSHSEtzO: the faster moving material (1.68 g) appeared with 10% ether; this was followed by unresolved mixtures (0.80 g ) eluted with 20% ether. The more polar component (0.3 g ) free from admixture was obtained wit,h 30yc ether. Seither material was crystalline. Examination of the nmr spectrum indicated t,hat both species weie di( triphenylmethyl) ethers on the basis of relative abundance arid po1aiit.y; the faster moving diether is tentatively assigned the 3',5'-ditrityl strricture of am-FU ( 6 ) . lp-~-Arabinofuranosy~-5-fluorouracil %'(?)-Acetate('I).-The material to which the 3',5'-ditrit8yl structure had been assigned (the less polar species 6, 1.52 g ) was dissolved in 5.0 ml of dry pyridine; AczO ( 2 . 5 ml) was added, and the solution was allowed to react' overnight at room temperature with protection against moist'ure. An equal volume of crushed ice was added, and all volatile components were stripped in vacuo. The residue was dissolved in 25 ml of S0Ci;. aqueous AcOH and heated for 1 hr a t 95' (oil bath). Again, t.he reaction mixture was concentrated to dryness, t,he residue was taken up in H20( 2 5 ml), and insoluble trityl alcohol was removed by filt,ration. The precipit,ate was washed with 23 ml of HZO and the combined aqueous filtrate %'as concentrated t o dryness. The residue was crystallized from acetoile-petroleum ether ( b p 30-60") to give 0.39 g of 7 (63%), mp 202-205'. An analytical sample had mp 206-208". Ana/.
527
(CIIHISFNnOT) C, H, N. ilbsorptioii peaks of spectra (uv, ir, nmr) were as expected. Similar treatment, of the more polar (minor) ditrityl derivative did not give rise to a crystalline material, although a homogeneous oil (tlc) with ail nnir spect.rum compat,ible with a monoacetate was obtained. lp-~-ArabinofuranosyI-5-fluorouraci~ 2',3'-Diacetate (9).-5'lIonotrit,yl derivative 5 (5.38 g, 10.7 mmol) was dissolved in 2.5 ml of anhydrous pyridine and 12..i ml of AcnO was added. Aft,er standing overnight at room temperature with t'he exclusion of moisture, an equal volume of crushed ice was added, and the resulting solution was concentrated in z~acuoto an oil. The latter was dissolved in 50 ml of 80% aqueous AcOH and heat,ed in a stappered flask for 1 hr at 95' (oil bath). The solution was again concentrated to a small volume in aacuo; H?O was added several times, and again the soliitioii wa.3 subjected to vacuum concentration to remove most of the acetic acid. The residue was suspended in 30 ml of H?O, the insoluble trityl alcohol was removed by filtration, and the solid was washed twice with 50 ml of H20. The combined aqiieous filtrates were concentrated to dryness, and the residue crystallized from lIeOH to give 0.9.7 g An analytical sample had mp of 9 (267,),19 mp 163-165'. 164-166"; absorption peaks of spect,ra (uv, ir, nmr) were as C, H, N. expected. Ana/. ( C18Hl,FN20e) lp-~-ArabinofuranosyI-5-fluorouracil 5'-Phosphate (4).Diacetat,e 9 (0.95 g, 2.74 mmol) and 13.i ml of a 1 Jf solution of pyridinnm cyanoethyl phosphate in pyridine (prepared according to Teiier23) were dried, by colicelitrating to dryness several times from dry pyridine, and dissolved in 2.5 ml of dry pyridine. K,K'-Dicyclohexylcarbodiimide (3.63 g, 27.4 mmol) was added, and the reaction mixture was stirred at room temperature for 24 hr, mokture being excluded. 4 t the end of that period t,he reaction was stopped by addition of 10 ml of HjO. After a further period of 18 hr in the refrigerator, the precipitated dicyclohexylurea was removed by filtration and washed with 100 ml of a 50% aqueous pyridine solution. The combined filtrate was extracted with four 100-ml portions of ether to remove excess reagent,. The aqueous phase was concentrated in uacuo, with two additions of EtOH to remove most of the pyridine, and dissolved in 2i.4 ml of a methanolic solution of 1 M NaOlIe and the resulting suspension was stirred at 60-6.7" (oil bath), under S2 in a vessel fitted with a reflux condenser. After 1 hr, the reaction mixtiire was cooled iii an ice bath and neutralized to pH 7 with anhydrous HCl in ethanol and concentrated to dryness in vacuo. The residrie as taken up in 50 ml 0.003 Jl triet~hylamnioiii~~ni bicarbonate, pH 7.5, and charged on a column of 1lEAE-cellrilose, 33 X 3 cm, equilibrated with the same biiRer. -4 gradient diliition system was applied: mixing chamber, 4 1. of 0.005 d l biifier as above: reservoir, 4 1. of the same, 0.3 M. Fractions of 20 ml were collected at the rate of 7,5('hr. The desired phosphate 4 was obtained from fractions 190-250, yield 57.50 OD units at , , ,A 268 mu ( p H i.5), or 0.72 mmol ( 2 6 5 from 8). The material was homogeneoils by paper chromatography (Rr 0.19 in system A ) and identical with that, obt,aiiied from enzymatic phosphorylation by spectra (nmr, nv), paper mobility, and enzymatic susceptibility to 5'-niicleotidase. (29) I n another experiment the lield was 32%.
