вход по аккаунту


Nucleosides and Nucleotides as Potential Therapeutic Agents.

код для вставкиСкачать
and, 3) the glutamine synthetase deadenylylating enzyme. Equation 4 shows the adenylylated and unadenylylated forms of glutamine synthetase, with their respective properties, as well as the enzymes which interconvert the two forms, and their effectors.
of the products of glutamine metabolism. In this form
glutamine synthetase activity is thus further diminished
by the accumulated nitrogenous metabolites, which
has as its result the lowering of the level of glutamine
in the cell.
Activated by glutamine. Inhibited by a-ketoglutarate and UTP
Unadenylylated glutamine synthetase
Deadenylylating enzyme
Adenylylated glutamine synthetase
Activated by a-ketoglutarate and UTP. Inhibited by glutamine and glutamate
More active form
Resistant to feedback inhibition
Mgz+ specific
It is clear from this scheme that conditions of nitrogen
starvation, with a correspondingly low nitrogen saturation ratio, lead to activation of the deadenylylating enzyme as well as to an inhibition of the adenylyltransferase, owing to the alterations in the relative
levels of glutamine and a-ketoglutarate. This results
in the formation of relatively unadenylylated glutamine synthetase, the form of the enzyme which is
intrinsically more active and less sensitive to feedback
inhibition. In this condition, ammonia will be converted into nitrogenous metabolites with high efficiency.
On the other hand, with high levels of the nitrogen
saturation ratio, as might be expected with a surfeit
of nitrogen-containing compounds,thedeadenylylating
enzyme would be inhibited and the adenylyltransferase activated. The result of this reciprocal enzyme
control would be to convert the glutamine synthetase
to the adenylylated form, which is less active and
additionally more susceptible to inhibition by several
Less active form
More sensitive to feedback inhibition
MnZ+ specific
The three-enzyme nitrogen regulatory system has as
its most obvious role the preservation of homeostasis in
cellular nitrogen metabolism; any shift away from the
“normal” nitrogen saturation ratio is accompanied
by adjustments in the three enzymes which tend to
restore the ratio toward the preexisting level. These
intricate controls have as an ancillary effect the preservation of cellular ATP, since both glutamine
synthesis and adenylylation of glutamine synthetase
consume ATP; in fact, the adenylylation-deadenylylation system without reciprocal controJ would be a
wasteful ATPase activity. The reason for the nucleotide controls in the adenylylation and deadenylylation phenomena is as yet obscure; a possible link
between glutamine synthesis and polynucleotide metabolism is suggested by the effects of ribonucleic acid
species on the deadenylylating system.
Received: October 17, 1969
[A 783 1E1
German version: Angew. Chem. 82, 689 (1970)
Nucleosides and Nucleotides as Potential Therapeutic Agents
By T. Y . Shen[*l
The present article summarizes recent progress in the study of nucleoside derivatives as
antiviral and immunosuppressive agents. A number of 5’-substituted 5‘-deoxy nucleosides
have been found to be permeable and nonincorporable antimetabolites of 5’-nucleotides.
N6-isopenfenyladenosine and analogs show certain promising immunosuppressive activities. Encouraged by the antibody-stimulating efJect of oligonucleotides, we have
developed a convenient synthesis of oligonucleotides using fully pro fected phosphorylated
intermediates. A group of tetradeoxyribonucleotides,dApdApdApdX, was prepared for
biological and physical evaluations. Nucleotide derivatives may prove valuable in the
treatment of’several immunological disorders.
1. Introduction
A brief review of some recent studies of nucleosides
and nucleotides in the field of chemotherapy and
immunology is presented below. The purpose of this
[*I Dr. T. Y . Shen
Merck Sharp and Dohme Research Laboratories
Rahway, New Jersey 07065 (USA)
discussion is to highlight several biomedical targets of
interest to medicinal chemists and to indicate the
therapeutic potentials of nucleosides and nucleotides 111.
[I] Adapted from presentations at the Symposium on Recent
Advances in Nucleoside Chemistry. 156th American Chemical
Society Meeting, Atlantic City, New Jersey, Sept. 11, 1968,
and a t the Gordon Conference o n Carbohydrate Chemistry,
New Hampshire, June 1969.
Angew. Chem. internat. Edit.
Vol. 9 (1970) 1 No. 9
During the past fifteen years, extensive chemical and
biological studies of nucleosides and some nucleotide
analogs have been stimulated mainly by their antitumor and antibiotic properties. The cytotoxic o r
bactericidal activities of these antimetabolites are
often attributable to the inhibition of enzyme systems
involved in the biosynthesis of nucleic acids, such as
nucleotide reductases, kinases, and polymerases, or
to the formation of fraudulent nucleic acids through
incorporation [21. Several nucleoside antibiotics such
as puromycin, blasticidins, and gugerotin inhibit
protein synthesis, either by chain termination or by
interference with the normal function of peptide
synthetase or ribosomes 133. However, since these components are also essential in the biosynthesis of protein
and nucleic acids in normal cells, the therapeutic value
of nucleoside antimetabolites are frequently diminished
by their potential toxicity.
The need for a better therapeutic index is particularly
important in the application of nucleoside analogs to
viral chemotherapy 141 o r metabolic diseases. Lack of
incorporation into normal cells (Le., lack of mutagenic effect) and a high degree of selectivity toxicity
are almost mandatory for any useful agent.
2. Inhibitors of DNA Viruses
2.1. 2'-Deoxyribo- and Arabinonucleoside Analogs
The first clinical application of a nucleoside as antiviral agent with an adequate therapeutic index was
demonstrated by a well-known thymidine analog,
S-iodo-2'-deoxyuridine ( 1 ) (idoxuridine), in the treatment of an eye infection, herpes keratitis 151. The
apparent safety in this case was accomplished by a
unique physiological phenomenon, namely, ophthalmic infections are topical and not unlike an in vitro
system. Since then several more potent DNA-synthesis inhibitors or thymidine antimetabolites, such as
trifluorothymidine (2) [61, cytosine arabinoside (cytarabine) (3) [TI, and its 5-fluoro analog (4) [*I, have
also shown similar efficacy. A homolog of thymidine,
5-ethyl-2'-deoxyuridine (Sa), was also reported to
have antiherpes activity [gal.
[2] J . A . Montgomery, Progr. Drug Res. 8, 431 (1965).
[3] R . E . Monro and K . A . Murcker, J. Molecular Biol. 25, 347
(1967); M . Yukioka and S . Morisuwa, J . Biochemistry (Japan)
66, 241 (1969).
[4] For summary see: Antiviral Substances. Ann. New York
Acad. Sci. 130 (1965); W. H . Prusoff. Pharmacol. Rev. 19, 209
151 H . E . Kaufman, E . L . Marfolu, and C . Dohlman, Arch. Ophthalmol. 68, 235 (1962).
[6] H. E. Kaufman and C. Heidelberger, Science (Washington)
145,585 (1964); R . A . Hyndiuk and H. E . Kaufman, Invest. Ophthalmol. 5, 424 (1966).
171 D . A . Buthala, Proc. SOC.exp. Med. 69, 115 (1964).
[8] R. Duschinsky, T . Gubriel, M . Hoffner, J . Berger, E . Titsworth, E . Grunberg, J . H . Burchenal, and J . J . Fox, J. med.
Chem. 9, 566 (1966); T . Y . Shen and W. V. Ruyle, US-Pat.
3 328 388 and unpublished results.
[8aI K. K . Gauri, G . Mulorny, and W. Schiff, Chemotherapy 14,
129 (1969), and literature cited there.
Angew. Chem. internat. Edit. J Vol. 9 (19701 1 No. 9
In our laboratory another analog, 5-methylamino-Z'deoxyuridine (MADU) ( 5 ) , was investigated [91. As
found by Visser and his co-workers, MADU is only
poorly active as a thymidine antagonist in the bacterial system f101. SurprisingIy, in tissue culture and
rabbit-eye systems MADU is as effective as idoxuridine against herpes simplex [111. The high degree of
selective toxicity against herpes virus was further
dramatized by its lack of inhibition against other
D N A and RNA viruses. It wassuggested that the
phosphorylated derivatives of MADU block the
utilization of thymidine 5'-triphosphate for DNA
synthesis. The selectivity of MADU is thus the consequence of an enhanced phosphorylating capacity of
infected cells 1123. Unfortunately the activity of MADU
against systemic herpes infections proved to be a
The in vivo activity of nucleosides is often limited by
their unfavorable distribution or metabolic characteristicsrl31. It was observed recently that the in vivo
immunosuppressive activity of 5'-adamantoyl cytarabine (6) is more potent and longer lasting than that
of the parent nucleoside (3) [141. Selective localization,
sustained hydrolysis, and resistance to enzymatic
degradation were suggested as possible explanations.
It would seem that similar derivatization of active
nucleosides with an in vivo cleavable group, a technique well studied in the production of other drugs, may
well expand the potential application of other nucleoside analogs.
R = I (Idoxuridine) ( I )
H (Cytarabine) (3)
More recently a purine nucleoside, adenine arabinoside
(7) [151, was described as a broad-spectrum anti-DNA
viral agent in vituo and in vivo"6J. I t is probably
superior to idoxuridine in the treatment of herpes
keratitis but its utility in systemic viral infections
remains to be established. In view of the recent impli191 T . Y . Shen, J . F. McPherson, and B. 0. Linn, J. med. Chem.
9, 366 (1966).
1101 D . W. Visser, S . Kubat, and M . Lieb, Biochim. biophysica
Acta 76, 463 (1963); S . Kubat and D. W. Visser, ibid. 80, 680
I l l ] M. M . Nemes and M. R. Hilleman, Proc. SOC.exp. Biol.
Med. 119, 515 (1965).
[12] R . W. Burg and M . M . Nemes, Federat. Meetings Abstract
1131 K . Gerzon and D . Kau, J. med. Chem. 10, 189 (1967).
1141 G. D. Gray, M . M . Mickelson, and J . A . Crim, Biochem.
Pharmacol. 18, 2163 (1969).
I151 W . W . Lee, A . Benitez, L . Goodman, and B. R . Baker, J.
Amer. chem. SOC.82, 2648 (1960).
1161 F. M . Schubel, Chemotherapy 13, 321 (1968).
cation of herpes virus in infectious mononucleosis,
Burkitt's lymphoma [171, and cervical carcinoma
safe and more effective compound is obviously very
( 7)
2.2. Search for Novel Structures
The potential role of soluble RNAs as regulators of
nucleic acid metabolism has often been discussed [191.
The significance of various minor o r abnormal
nucleotides present is particularly intriguing. In our
study of pyrimidine nucleosides as novel antimetabolites it was noted that several thiopyrimidine
nucleosides, such as 4-thiouridine (8) 1201, 2-thiocytidine (9), and the 5-substituted derivatives (10) [211
and ( 1 1 ) [221, have been detected in t-RNA. Alteration
of 3 5 s incorporation into t-RNA has also been observed after a phage infection 1231.
2-Thiocytosine arabinoside (12) was synthesized in
our laboratory by the ring-opening of a cyclo-
nucleoside [241, but to our disappointment it does not
possess any significant antiviral properties.
So far, anti-DNA viral activity seems to be mostly
associated with pyrimidine analogs with 2'-deoxy-~ribose and D-arabinose moieties. A group of compounds with pyrimidine bases such as uracil, thymine,
5-bromouracil, and cytosine attached to modified
pentofuranoses such as 3-amino-2-3-dideoxyribose
(13), 2,3-anhydroribose (14), 2,3-dideoxyribose ( I S ) ,
or its 2,3-didehydro analog (16), and several hexopyranoses are all inactive 1251. Recently, the 2',3'-didehydro analog of 5-fluorouridine was found to be an
anti-tumor agent [*61, but its antiviral activity has not
been described.
(lo), R
(I]), R
(171 G . Henle, W . Henle, and V. Diehl, Proc. nat. Acad. Sci.
USA 59, 94 (1968); L . N . Chessim, P . R . Glade, J . A . Kasel,
H . L. Moses, R . B . Heberman, and Y . Hirschaut, Ann. intern.
Med. 69, 333 (1968).
[18] W . E . Rawls, W .A . F. Tompkins, M . E . Fiqueroa, and J . L .
Melnick, Science (Washington) 161, 1256 (1968).
[19] M . J . Robins, R . H . Hall, and R . Thedford, Biochemistry 6 ,
1837 (1967).
[20] M . N . Lipsett, J. biol. Chemistry 240, 3975 (1965).
[21] J . Carbon, H . David, and M . H . Strrdier, Science (Washington) 161, 1146 (1968).
[22] L. Baczynsky, K . Biemann, and R . H . Hall, Science (Washington) 159, 1481 (1968); L . Baczynsky, K . Biemann, M . H . Fleysher, and R . H . Hall, Canad. J . Biochem. 47, 1202 (1969).
1231 S. B. Weiss, Wen Tah Shu, J . W . Foft, and N . H . Scherberg
Proc. nat. Acad. Sci. USA 61, 114 (1968).
A group of branched chain nucleosides, e.g., 2'-Cmethyl and 3'-C-methyl homologs of adenosine and
cytidine, was synthesized recently 127,281. Compared
[24] W . V. Ruyle and T . Y. Shen, J. med. Chem. IO, 331 (1967).
(251 Unpublished results from our laboratories.
1261 T . A . Khwaja and C . Heidelberger, J. med. Chem. 10, 1066
[27] E . Walton, S. R . Jenkins, R . F. Nutf, M. Zimmermann, and
F. W . Holly, J. Amer. chem. SOC.88, 4524 (1966).
[28] E . Walton, F. W . Holly, S . R . Jenkins, R . F. Nutf, and
M . M . Nemes, J. med. Chem. 12, 306 (1969).
Angew. Chem. internat. Edit. / Vol. 9 (1970)
1 No. 9
with 3'-deoxyadenosine (cordycepin), both 3'-Cmethyladenosine (18) and 3'-C-methyIcytidine (19)
are much less toxic to cell cultures L291. However, both
showed interesting activity against vaccinia infection
in mice when administered at doses of 1-2 mg and are
superior to t h e standard drug, Marboran. The corresponding 2'-C-methyl analogs are much less active [281.
The active site of R N A polymerase contains a histidyl
residue 1331 which is essential for chain elongation.
Analogous to the active site in nucleoside diphosphokinases [341, the histidyl group may form a phosphorylimidazole intermediate (20), with the nucleoside triphosphate.
Speculatively, a concerted activation (step 1) and
transfer (step 2) sequence may be visualized as shown
in Scheme 1.
3. Search for Inhibitors of RNA Viruses
3.1. Structure Information of RNA Polymerase
In the search for anti-RNA viral agents, our attention
was directed to the unique synthesis of viral RNA,
which involves a single-stranded RNA template, a
double-stranded replicative intermediate, and a viral
specific polymerase. The complexity of viral replication has generated much speculation and many
hypotheses 1301. It would seem that novel reversible o r
irreversible inhibitors derived from nucleosides, in
addition to their therapeutic value, might also be useful as molecular probes of the active site in polymerase
and help to elucidate its mechanism. Current studies
of R N A polymerases indicate that the incoming
nucleoside triphosphate may interact first with the
active site to form a reactive nucleoside-5'-phosphorylenzyme intermediate which then participates in
coupling with the 3'-hydroxy group of the growing
polymer. The sequence of reaction is similar to that
in polynucleotide ligase [311 (an enzyme whose activity
is changed significantly after viral infection and which
may turn out to be a chemotherapeutic target also) 1321.
As a model of this active intermediate we have synthesized the crystalline compound (21) [351. The site of
phosphorylation is tentatively assigned as N-3, based
on the observation that the N-3-phosphoryl derivatives
3.2. Models of the Enzyme Intermediate
Scheme 1 .
or DPX
[29] H . T . Shigeura, unpublished observations.
[30] Nature (London) 219, 675 (1968).
1311 2. W . Hall and I . R . Lehman, J. biol. Chemistry 244, 43
[321 J . Sambrook and A. J . Shatkin, J. Virology 4 , 719 (1969).
Angew. Chem. infernat. Edit. Vol. 9 (1970)
/ No. 9
[33] A . Ishihama and J . Hurwitz, J. biol. Chemistry 244, 6680
[34] P . L. Pedersen, J. biol. Chemistry 243, 4305 (1968).
[3S] K . H . Boswell and T . Y. Shen, unpublished.
68 1
of histidine are generally more stable than those phosphorylated at N-1 [361.
Unlike many adenylic acid derivatives, which have
ORD spectra similar to that of 5’-AMP with a
negative Cotton effect at 277 nm, compound (21)
gave an anomalous ORD spectrum with a positive
Cotton effect near 270 nm. Whether the reversal of
Cotton effect is indicative of a conformational change
due to the intramolecular interaction of the imidazole
moiety and adenine chromophore remains speculative,
but any correlation of the physicochemical properties
with those of the active intermediate of polymerase
would certainly be worth investigating. A similar reversal of the Cotton effect is found for the 5’-adenylyl
derivative of imidazole, pyrazole, and 2-hydroxypyridine 1351.
The possibility of using these analogs of an active
intermediate to interfere with the enzyme action are
still being investigated. It is assumed that in the normal
course of polymerization the formation of the active
intermediate (imidazole-p-A) is followed by transfer
of p-A to the growing chain of the polynucleotide.
This would require either a conformational change of
the active site, e . g . , prompted by the elimination of
the charged pyrophosphate group, or a translocation
of the active intermediate itself. One could imagine
that these analogs may well interfere with the transfer
step. Classical antimetabolites of the substrate ATP,
with their charged triphosphate-like side chain, will
probably interfere only with the activation step and
the formation of the active intermediate.
erase substrate [391. These 5’-substituted nucleosides
probably permeate easily into cells. In addition, they
are attractive for several important reasons. Firstly,
with their 5’-hydroxyl group replaced by a substituent
they are totally nonincorporable, thus fulfilling our
safety requirement mentioned above. Secondly, unlike
most nucleoside antimetabolites, any inhibitory action
of a 5‘-substituted nucleoside should not depend upon
its prior enzymatic conversion into a nucleotide, and
resistance development is therefore less likely 1403.
Furthermore, as recently shown by 5’-deoxy-5’methylthioadenosine [411, 5’-substituted adenosine derivatives might be more resistant to catabolic enzymes
such as adenosine deaminase and purine nucleoside
phosphorylase. Although unknown at the beginning
of our study, the feasibility of using a 5‘-substituted
nucleoside as kinase or polymerase inhibitors was
strongly indicated by the recent disclosure of 5‘-deoxy5‘-fluorothymidine (24) as a competitive inhibitor, not
of thymidine to its kinase, but of thymidylic acid to
thymidylate kinase [421. Another example of a 5’substituted nucleoside analog is the naturally occurring
antibiotic nucleocidin (2s) 1431 whose structure was
elucidated recently.
3.3. 5‘-Substituted Nucleoside Analogs as
Nucleotide Equivalents
The best known classical antimetabolites of nucleotides
are their methylene phosphonate isosteres, e.g., compounds (22) [371 and (23) [381. In both cases a n oxygen
linkage is replaced by a methylene group that is noncleavable under physiological conditions.
Depending on which oxygen atom is replaced by the
methylene group, these analogs are either substrates
or inhibitors 1371. However, their practical applications
are obviously restricted by their poor permeability
properties. To circumvent this barrier, we decided to
investigate a group of nucleoside derivatives bearing
a polar organic moiety a t the 5’-position which might
mimic the triphosphate-Mg2+side chain of the polym[36] D . E. Hultquist, Biochim. biophysica Acta 153, 329 (1968).
[371 T . C. Meyers, K . Nakamura, and A . B. Danielsaheh, J. org.
Chemistry 30, 1517 (1965).
[38] G . H . Jonesand J . G. Moffat, J. Amer. chem. SOC.90,5337
Our endeavor to construct an organic equivalent of
the 5’-triphosphate-Mg2+ complex was much hampered by the uncertainty on the configuration of
nucleotides such as the ATP-MgZ+ complex. In general, a cyclic chelating structure formed by the and y
phosphates and Mgz+ is favored, but the possible
1391 T . Y. Shen, Abstracts of Papers, 154th Amer. Chem. SOC.
Meeting, Sept. 1967, Chicago, Ill., p. 29.
[40] B. R . Baker and P . M. Tanna, J. pharmac. Sci. 54, 1774
1411 A . E . Pegg and H . G . Williams-Ashman, Biochem. J. 11-7,
241 (1969).
1421 P . Langen u. G. Kowollik, Europ. J. Biochem.6, 344 (1968).
[431 G . 0. Morton, J . E. Lancaster, G . E . Van Lear, W. Fulmor,
and W . E. Meyer, J. Amer. chem. SOC.91, 1535 (1969).
Angew. Chem. internat. Edit.
/ Vol. 9 (1970) 1 No. 9
involvement of a ring nitrogen in the coordination
remains to be established. Very recently structures
(26) involving the participation of N7, instead of the
6-amino group, have been suggested on the basis of
relevant X-ray data 1441. The three-dimensional structure of ATP in the hydrated disodium salt, with the
phosphate chain folded back towards the purine base,
was reported by another group (451.
R = H or Y=NH-CO-NHz, NH-C(NH)-NH2,
R-R = isopropyIidene NH-CS-NH2, NH-C(NH)-NHC4Hg
3.4. Derivatives of S'-Amino-5'-deoxy Nucleosides
Our initial plan was to explore various nucleoside
derivatives which have a cyclic or chelating structure
at the 5'-position to mimic the triphosphate side chain,
and the 5'-amino-5'-deoxy
and S'azido-5'-deoxy
derivatives of adenosine (27) and (28), respectively [461,
were chosen as key intermediates for their synthesis.
These selections turned out to be particularly fortuitous since both compounds exhibited antiviral
activity per se; inhibition of parainfluenza I11 was
demonstrated 1471 at 8-30 tJ-g/mlin tissue cultures with
therapeutic indices ranging from 16 to 128, depending
upon the severity of the infection. Encouraged by
these findings, we synthesized a number of simple
derivatives from the versatile 5'-tosylate (29) by
similar nucleophilic displacement reactions. To minimize formation of the cyclonucleoside (30), the N6formyl and 6-methylthio derivatives were employed in
some cases.
The protected 5'-arnino-S'-deoxyadenosine (32) was
phosphorylated with dibenzyl phosphite and N-chlorosuccinimide in methylene chloride to the phosphoramidate (33). The isopropylidene group was first
removed by dilute formic acid to give (34), which, on
(34), R = H
hydrogenolysis in the presence of sodium hydroxide
afforded a product, believed to be the desired 5'-aza
AMP, that was too unstable for cyclization to the
= N3, NH2, CH3NH, (CH&N, CH3-C(O)S, -N=C=NH,
Several guanidine and urea derivatives were readily
prepared from the cyanamide precursor (31).
I441 M . Sundaralingam, Biopolymers 7, 821 (1969).
[45] 0.Kennard, N . W . Isaacs, J . C . Coppola, A. J . Kirby,
S . Warren, W . D . S . Motherwell, D . G . Watson, D . L. Wampler,
D . H . Chenery, A. C. Larson, K . A. Kerr, and L . R . DiSanseverino, Nature (London) 225, 333 (1970).
[46] W . Jahn, Chem. Ber. 98, 1705 (1965).
I471 M . M . Nemes, unpublished observations.
Angew. Chem. internat. Edit. J Vol. 9 (19701 No. 9
desired 5’-aza-3’,5‘-cyclic AMP. A similar observation
regarding the stability of (35) was reported recently 1481.
A / C
In the 2‘-deoxyribonucleoside series, 2’-deoxyadenosine, a known antimetabolite, was sulfonated to give
a mixture of mono- and di-tosylates. Treatment of
their formyl derivatives (36) and (37), respectively,
with sodium azide gave the corresponding azido
derivatives (38) and (39). The former was reduced to
5‘-amino-2‘-5’-dideoxyadenosine(40) [491.
3.5. 5‘-Substituted Arabinosyl Nucleosides
The importance of arabinosyl nucleosides in the
therapeutic field has been amply demonstrated by
cytosine-, adenine- and 6-methylthiopurine arabinosides. Cytosine arabinoside was readily transformed
to its 5‘-amino analog (41).
itself [51J. The 4-keto group in (45) was converted to
the 4-amino via the methylthio intermediate, and the
protecting groups were removed to give (41).
The arabinosyl halide (44) was condensed with adenine
and then deblocked to give 5‘-amino-5‘-deoxyadenine
arabinoside (46). The 6-mercaptopurine analog (47)
was obtained in a small yield from the protected 6chloropurine precursor by treatment with thiourea. In
this case the removal of protecting groups was complicated by the formation of disulfide and other decomposition products.
R’ = NH,, C 1
R’= NH2,C1
In order to develop a versatile route to study 5‘substituted arabinosyl nucleosides, we first synthesized
methyl 5‘-azidoarabinofuranoside (42) as a potential
intermediate. Unexpectedly, attempted mild acid
hydrolysis yielded a new crystalline compound which
was found to be the piperidone (43) [501.
On the other hand, when a phthalimido group was
placed at the 5’-position, the protected arabinosyl
halide (44) could be prepared readily from the corresponding arabinosyl p-nitrobenzoate and used in the
standard Hilbert-Johnson reaction. The $-nucleoside
(45) was formed as the predominant product from
the a-chloride. The stereospecificity is reminiscent of
our earlier experience with 2,3,5-tri-O-benzylarabinosyl halide in the synthesis of cytosine arabinoside
[48] B . Jastorffand H . Hettler, Tetrahedron Letters 1969, 2543.
[49] An independent synthesis of this compound was recently
reported: M . G . Stout, M . J . Robins, R . K . Olsen, and R . K .
Robins, J. med. Chem. 12, 658 (1969).
[50] W . V. Ruyle, unpublished observations.
(46/, R2 = NH,
(47). RZ = SH
The P-configuration of these purine arabinosides was
corroborated by the similarity of their ORD spectra
with that of adenosine. Previous workers have shown
that the configuration of the 2‘-hydroxyl group has
no significant effect on the ORD of these nucleosides.
3.6. Biological Properties of 5’-Substituted Nucleosides
The new analogs exhibited moderate to good cytotoxicity vs. HeLa cells at concentrations of 1 millimole/] 1291. The 5’-azido-5‘-deoxy derivatives of both
[ S t ] T . Y.Shen, H . M . Lewis, and W . V . Ruyle, J. org. Chemistry
30, 835 (1965).
Angew. Chem. internat. Edit.
/ Vol. 9 (1970) J NO. 9
adenosine and 6-methylthiopurine riboside are highly
toxic at this level. Several analogs, e.g., the 5’-amino,
5’-azido, 5‘-ureido, and 5‘-ethoxycarbamyl derivatives
of adenosine, were shown to inhibit the conversion of
adenosine to its phosphorylated derivatives in ascites
cells to cn. 60 ”/, at concentrations of 0.9 mmole/l r291;
however, no significant antiviral activity was uncovered. The inhibition of adenosine utilization by
these compounds is reminiscent of the inhibition of
thymidine utilization by 5’-deoxy-5’-fluorothymidine
(24) mentioned earlier.
As part of our systematic structure modifications, the
5‘-azido derivative (49) of 6-isopentenyladenosine
(48) (6-IPA) L521 was also synthesized. This derivative
is again a nonincorporable antimetabolite. Yet, like
6-IPA, it inhibited HeLa cells to almost 100 ”/, at a
concentration of 1 mmole/l, although its potency was
somewhat less at lower concentrations.
(481, K = OH ( 6 - I P A )
(49), R = N,
4. Immunological Effects of Nucleosides and
4.2. Properties of Cytokinin Derivatives
Kinetin riboside (52), 6-IPA, and other cytokinin
derivatives represent a new group of immunosuppressants. The primary effect of 6-TPA was suggested as
interference with R N A and protein synthesis [Sjl. It
has recently been observed that, depending on the
concentration and on the particular stage of the cell
cycle, 6-IPA has a biphasic effect on the transforniation and mitosis of human lymphocytes treated with
phytohemagglutinin. This biphasic response is reminiscent of the antibody stimulating o r inhibitory
effect of another RNA synthesis inbibitor, 6-MP, in
several in viva systems (561.
6-IPA and cytokinins also exert an indirect influence
on certain humoral immune responses mediated by
nucleic acids. The participation of two or more cell
types, macrophages and lymphocytes, in the primary
response has been demonstrated 157,581. In the case of
certain soluble antigens, a 4 S RNA or a RNA-protein
complex is released by macrophage after phagocytosis
of the antigen. This RNA-containing material, the
nature of which is still being elucidated, is possibly
involved in the sensitization of lymphocytes 1591.
Coincidentally, nucleic acid digests or oligonucleotides
have been shown to have an adjuvant effect in enhancing the phagocytosis of antigen by macrophage
and in stimulating the production of antibodies by
sensitized lymphocytes (Fig. 1)1601. For reasons not yet
understood, the antibody stimulating effect is inhibited
by 6-IPA, kinetin riboside (521, and other analogs [611.
stiniuIation b y oIigonucIeotides
kinetin analogs
4.1. Immunosuppressive Nucleosides
Our interest in 6-IPA analogs was further stimulated
by their potential application as immunosuppressive
agents. Nucleoside antimetabolites such as 6-mercaptopurine, imuran, and cytarabine (3) have been
used in clinical treatment of various immunological
disorders. 6-Mercaptopurine arabinoside (50) and 6methylthiopurine riboside (51) and its periodate
oxidation product have also been studied in the
laboratory 1531. Using the hemagglutinin response to
sheep erythrocytes as a model for humoral antibody
production and graft versus host reactions as representative cell-mediated immune reactions, the
mechanism of action of these immunosuppressants is
gradually being elucidated 1531. It is instructive to
realize that these D N A synthesis inhibitors are very
effective in suppressing the proliferation of sensitized
lymphocytes and cell-rnediated immunity, but the
biochemical blockades imposed are easily overcome.
Humoral antibody response is generally not inhibited
by these compounds 1541.
[521 K . Biemann, S . Tsunakawa, K. Sonnenbichler, H. Feldmann, D . Diifting, and H. G. Zarhair, Angew. Chem. 78, 600
(1966); Angew. Chem. internat. Edit. 5. 590 (1966).
[531 J . P. Bell, M . L . Faures, G . A . LePnge, and A . P . Kimball,
Cancer Res. 28, 782 (1968).
[541 R . H. Gisler and J . P. Be“, Biochem. Pharmacol. 18,2115,
2123 (1969).
Angew. Chem. infernat. Edit. j Vol. 9 (1970) J No. 9
HNA - protein
+ lymphocyte
Fig. 1. Effect of cytokinin o n antibody production.
Cytokinins are usually alkylated purine derivatives 1621,
but a pyrimidine cytokinin, 6-methyluracil (pseudothymine) (53), was described recently 1631. Jt would be
of interest to compare the immunosuppressive activity
of pseudothymine nucleosides 1641.
[55] R . C. Gallo, J . Whang-Peng, and S . Perry, Science (Washington) 165, 400 (1969).
I561 E . Gabrielson and R . A . Good, Adv. Immunology 6, 148
I571 M. Fishman and F. L. Adler, J . exp. Med. 117, 595 (1963).
[58] M . Fishman, Annu. Rev. Microbiol. 23, 199 (1969).
[59] A. A. Gottlieb, V . R . Glisin, and P . Doty, Proc. nat. Acad.
Sci. USA 57, 1849 (1967); A . A. Gottlieh and D . S. Straus, J.
biol. Chemistry 244, 3324 (1969).
[60] W . Braun and W . Firshein, Bacteriol. Rev. 31, 83 (1967).
I611 O.J. Plescia and W . Braun: Nucleic Acids in Immunology.
Springer-Verlag, Wien-New York 1968, p. 347.
[621 J . P. Helgeson, Science (Washington) 161, 974 (1968).
[631 B. I. Pozsar and Gy. Matolcsy, Life Sciences 7 , 699 (1968).
I641 M . W. Winkley and R . K. Robins, J. org. Chemistry 33,
2822 (1968).
68 5
Another alkylated purine, 1-methyladenine (54), is a
potent meiosis inducer which stimulates ovulation in
star fish at a concentration of 0.02 pg/ml[651. The
corresponding 1-ethyl homolog is much less active.
We have noted that I-isopentenyl adenine (55), an
intermediate in the synthesis of N6-isopentenyladenine,
inhibits thymidine incorporation in the HeLa cells at
1 mmole/l but increases thymidine and uridine incorporation in phytohemagglutinin-stimulated rat
spleen cells at 10 pg/ml[661.
Both 6-IPA (48) and its 2-methylthio derivative have
been found to occupy positions adjacent to certain
anticodon regions in t-RNA 167,681. The special location of these bases and the effectiveness of hydrophobic
N6 side chains have led to the suggestion that the
cytokinin in t-RNA functions in the codon-anticodon
interaction, possibly with an increasing ribosomal
binding by the N6-substituents. Whether the regulatory
effect of exogenous cytokinins is achieved by the incorporation into t-RNA per se o r by an indirect influence on the “pool” of cytokinin-containing t-RNA
species remains to be elucidated. The asymmetric N6alkyl side chains in several cytokinins preferentially
adopt the (S)-configuration
The effects of these nucleoside analogs on mammalian
cells have received much attention recently. Differential toxicity to certain leukemic cells and normal
lymphocytes has been observed with a group of N6alkylated adenosines [70-721. Obviously an extension
of this kind of inhibitory action to immunological
disorders would be of considerable interest.
(511,R = SCH,
(52),R = NHCH, 0
L65] H . Kanatani, H . Sirai, K . Nakanishi, and T . Kurokowa,
Nature (London) 221, 273 (1969).
[66] H . T . Shigeura and T . L . Feldbush, unpublished observations.
[67] W. J . Burrows, D . J. Armstrong, F. Skoog, J . M . Hecht,
J.T. A . Boyle, N.J.Leonard, and J. Occolowitz, Science(Washington) 161, 691 (1968).
[681 D . J . Armstrong, W . J . Borrows, F. Skoog, K . L . Roy, and
D. Soll, Proc. nat. Acad. Sci. USA 63, 834 (1969).
[69] K . Koshimizu, A . Kobayashi, T . Fukita, and T. Mitsui,
Phytochemistry I , 1989 (1968).
[70] N.J. Leonard, S. M . Hecht, F. Skoog, and R . Y. Schmitz,
Proc. nat. Acad. Sci. USA 59, 15 (1968).
5.1. Adjuvant Effect of Oligonucleotides
In our laboratory we were also intrigued by the enhancement of host resistance to infection by oligonucleotides C601. In the original immunological experiments an antibody-stimulation effect was observed
with a gross mixture of DNA digest and the optimal
size of the oligomers from DNA digest was estimated
to be tri- to hexanucleotides. Our aim then was to
delineate the active structural features of oligomers,
and, if possible, to shed further light on the mechanism
of this stimulating effect. Base sequence, charge distribution, molecular conformation, nature of backbone, and stability to nuclease cleavage were some of
the parameters to be evaluated. The cellular uptake of
oligonucleotides and the site of action, whether membrane or cytoplasmic, also required clarification (see
Fig. 2).
Fig. 2. Immune stimulating effect of oligonucleotides.
1. Enhancement of phagocytic activity.
2. Reduction of induction period in antibody production.
3. Increase of deoxynucleotide kinase activity (dCMP, dGMP).
4. Protection against bacterial infection in vivo.
Active structures: Oligonucleotides (n = 3-6)
Poly C , Poly A (Poly C
+ Poly I).
5.2. Synthesis of Oligo(2’-deoxynucleotides)
As an initial step, a group of tetradeoxynucleotides of
the general type dApdApdApdX (56) with dX = dA,
dG, dI, T, and dC were synthesized by a modified
procedure which incorporates the advantages of several well known techniques [73,74a,74bl.
5. Oligonucleotides
N6 - Benzoyl-5’- 0-monomethoxytrityl-2‘- deoxyadenosine (57) was used as starting material. Treatment
with trichloroethyl phosphate and mesitylenesulfonyl
chloride in pyridine gives the 3’-phosphorylated
derivative (58). Condensation with the 5‘-hydroxyl of
N6-benzoyl-2‘-deoxyadenosinewas also carried out in
pyridine in the presence of triisopropylbenzenesulfonyl
chloride to give the dinucleotide (59). The sequence
of 3‘-phosphorylation and 5‘-condensation was repeated to give the trinucleotide (60) and this in turn
was condensed with appropriately protected deoxyadenosine, deoxyguanoside, and deoxycytidine. 2’Deoxyinosine and thymidine were used directly. The
use of fully protected intermediates enabled us to
employ silica gel column chromatography for their
isolation and to obtain 10-20 mg of products without
[71] M . H . Fleysher, M . T . Hakala, A . Bloch, and R . H . Hall,
J. med. Chem. I I , 717 (1968).
[I21 M . H . Fleysher, A . Bloch, M . T . Hakala, and C. A . Nichol,
Abstracts of Papers, 156th Amer. Chem. SOC. Meeting, Sept.
1968, Atlantic City, N. J. Medi 25.
[73] H . G . Khorana et al., J. Amer. chem. SOC. 89, 2148, 2185,
2195 (1967).
[74] a) F. Eckstein and J. Rizk, Chem. Ber. 102, 2362 (1969);
b) R . L. Lersinger and K . K . Ogilvie, J. Amer. chem. SOC.89,
4801 (1967).
Angew. Chem. internat. Edit. 1 Yol. 9 (1970)
No. 9
too much difficulty. The protecting groups in the
tetranucleotide (61) were removed sequentially by
treatment with Zn/Cu couple, dilute ammonia, and
aqueous acetic acid. The selection of trichloroethyl as
our phosphate-protecting group also conferred versatility to this approach for further structure modifications. The scope of this synthetic sequence was
recently extended to the oligoribonucleotide series in
which the 2’-tetrahydropyranyl was used as protected
intermediate [751. Conversion of our oligodeoxyribonucleotides to cyclic (circular) analogs in order to
increase their resistance to nucleases and to enhance
cellular uptake is still under study; we are particularly interested in the physicochemical properties of
the products.
(56),X = A, T, G, C, I
5.3. Potential Applications of Oligonucleotides
O=P - OCHzCC1,
0-CH, ARz
0 - P -OCH,CCl,
O = P- OCH, CC 1,
1. Z n - C u / D M F ( 5 0 % )
2. NHiOH
0-CH, X
Angew. Chern. internat. Edit.
/ Vof.9 (1970) / No. 9
> (56)
Another potential application of nucleotide derivatives
lies in the area of autoimmune diseases, such as
systemic lupus erythmatosus (SLE) and other chronic
degenerative disorders, which are characterized by the
production of anti-nuclear antibodies directed against
nucleic acids and nucleoproteins. A variety of antibodies to native DNA, single-stranded DNA, and
even double-stranded R N A are found in patients with
SLE [76,771. These antibodies react with nucleic acids
to form antigen-antibody complexes which cause inflammation along the basement membrane of renal
glomerulus leading to the development of glomerular
lesions. On considering the formation of pathogenic
antigen-antibody complexes one is reminded of the
fact that the binding of antigen with antibody is
analogous to enzyme-substrate interactions. It is well
established that the interaction of polysaccharide and
its antibody can be inhibited by chemical analogs of
the determinant group, such as oligosaccharides 1781.
Similarly, a partial inhibition of SLE sera by tetra- or
pentanucleotides was demonstrated several years
ago [791. With more precise information on the nature
of binding involved it is conceivable that one may find
oligonucleotide derivatives not only with higher affinity but also capable of cross-reacting with a large group
of SLE sera. The formation of soluble antigen-antibody complexes between oligonucleotide analogs and
anti-DNA antibodies would presumably minimize
immunological lesions and possibly facilitate the induction of tolerance (see Fig. 3) [SO].
The concept of using derivatives of oligonucleotides o r
much simplified “structure equivalents” to mimic or
to compete with biologically active polynucleotides
should be applicable to other important problems of
[75] T . Neilson, Chem. Commun. 1969, 1139.
[76] D . Koffler, R . I . Carr, V . Angello, T . Fiezi, and H . G . Kunkel, Science (Washington) 166, 1649 (1969).
1771 P . H . Schur and M . Monroe, Proc. nat. Acad. Sci. USA 63,
1108 (1969).
[78] E. A. Kabat: Structural Concepts in Immunology and
Immunochemistry. Holt, Rinehart and Winston, New York
[79] L . Levine and B. D. Stollar, Progr. Allergy 12, 161 (1968).
1801 A . R . Boyns and J . Hardwicke, Immunology 15,263 (1968).
+ DNA (Ag
- -iit‘
TsOCH, T h
+ Ab-Ag‘
Fig. 3. Pathogenesis in systemic lupus erythmatosus.
medicinal interest such as interferon induction, viral
RNA polymerase inhibition, specific t-RNA inhibition, and episome transfer. I n each case one may
assume that the interaction of nucleic acid with a
protein molecule o r a receptor involves only one
region o r several regions of oligonucleotides and that
the specificity is determined by the base sequence and
secondary structure of those regions. The feasibility
of this kind of approach is clearly indicated by recent
sequence determinations of QP and R17 phage RNA.
The base sequence of QB phage was determined from
both the 3’- [81,82J and 5’-terminals @31, and no less
than 175 nucleotides from the 5’-terminus with a welldefined loop-like secondary structure have been elucidated @31. The identification of a specific binding
region should enable us to design chemical structures
either to compete for the protein binding site or to
interact with the oligonucleotide region itself by means
of intercalation, base pairing, o r formation of other
hydrogen bonds.
6. Novel Nucleoside Derivatives
6.1. “Double-Headed” Nucleosides
To expIore nucleoside analogs which might show a
selectivity for certain base-sequences on interaction
with nucleic acid, we have utilized the versatile 5’-0tosyl and 5‘-amino-5‘-deoxy derivatives of nucleosides
described above to synthesize novel ribonucleosides
and 2’-deoxyribonucleosides which possess a second
purine or pyrimidine moiety at the 5’-carbon C84351.
It is hoped that with the proper selection and spacing
of the two bases, one may utilize one base for WatsonCrick type pairing and the other to intercalate between
the next two adjacent nucleotides in a polynucleotide.
[81] U. Rensing a nd I . T. August, Nature (London) 224, 853
1821 H . L . Weith and P. T. Gilham, Science (Washington) 166,
1004 (1969).
[83] M . A. Billiter, J . E. Dahlberg, H . M. Goodman, J . Hindley,
and C . Weissmann, Nature (London) 224, 1083, 1055 (1969).
[84] R . Fecher, K . H. Boswell, J . Wittick, and T. Y. Shen, J.
Amer. chern. SOC.92, 1400 (1970).
1851 K . H . Boswell and T. Y . Shen, unpublished.
The specificity of these interactions will be the basis
for selective binding of three nucleotides by one
double-headed unit. The nucleoside (63) gave an
unusual CD spectrum 1841, presumably due to intramolecular interactions between the two proximate
bases. These compounds are useful models for the
study of base-base interactions. The pronounced
Cotton effect may also serve as a convenient marker in
following the intermolecular interactions between
these compounds and polynucleotides.
7. Conclusion
The medical applications of nucleotide derivatives have
not received much attention in the past partly because
of the difficulties in chemical synthesis and partly
because of their poor absorption and cellular permeability characteristics. However, recent developments in
synthetic methods and isolation techniques have
brought oligomers within the sphere of medicinal
chemical explorations. New discoveries, such as the
use of double-stranded polynucleotides as interferoninducers in viral chemotherapy 186,871, have broadened
the traditional concept of using only monomeric
drugs. The growing knowledge of membrane receptor
sites and membrane enzymes such as ATPase in the
regulation of cellular activities also provides new extracellular targets for drug action. With the structure
elucidation of nucleic acids and the understanding of
their interaction with protein receptors, further advances in the application of nucleoside and nucleotide
derivatives as novel therapeutic agents are certainly to
be expected.
Received: April 20, 1970
[A 779 IEI
German version: Angew. Chem. 82, 729 (1970)
[86] G . P. Lampson, A . A . T j f e l l , A . K . Field, M . M . Nemes,
and M . R. Hilleman Proc. nat. Acad. Sci. USA 58, 782 (1967).
[87] J . H . Park and S . Baron, Science (Washington) 162, 811
Angew. Chem. internat. Edit.
VoI. 9 (1970) No. 9
Без категории
Размер файла
966 Кб
potential, nucleotide, therapeutic, agenti, nucleoside
Пожаловаться на содержимое документа