Intracellular trapping of cyclosal-pronucleotides by enzymatic cleavage

Article abstract of DOI:10.1080/15257770701503886

A new synthesis for cycloSal-pronucleotides bearing enzymatically cleavable triggers is presented. This trigger is introduced to trap the pronucleotide inside cells. The general concept and hydrolysis data in different media are discussed. Copyright Taylo

Full text of DOI:10.1080/15257770701503886

Nucleosides, Nucleotides, and Nucleic Acids, 26:827–830, 2007
Copyright ꢀ Taylor & Francis Group, LLC
C
ISSN: 1525-7770 print / 1532-2335 online
DOI: 10.1080/15257770701503886
INTRACELLULAR TRAPPING OF CYCLOSAL-PRONUCLEOTIDES BY
ENZYMATIC CLEAVAGE
H. J. Jessen, V. Tonn, and C. Meier
Department of Chemistry, Organic Chemistry,
2
Faculty of Science, University of Hamburg, Martin-Luther-King-Platz, Hamburg, Germany
A new synthesis for cycloSal-pronucleotides bearing enzymatically cleavable triggers is presented.
2
This trigger is introduced to trap the pronucleotide inside cells. The general concept and hydrolysis
data in different media are discussed.
Keywords cycloSal-nucleotides; pronucleotides; antiviral nucleosides; nucleoside ana-
logues; prodrugs
INTRODUCTION
Nucleoside analogs are widely applied in antitumor and antiviral
therapy. However, in many cases the active form is not the nucleoside itself
but its triphosphate, which is then incorporated into DNA. Sometimes
the cellular or viral kinases do not accept the analogs as substrates or the
phosphorylation proceeds at low rates, leading to a decrease or loss of
antiviral activity. To circumvent these problems, it would be favorable to
apply at least the monophosphates, but these polar compounds do not
penetrate cell membranes and are rapidly degraded in plasm. Another
possibilty of enhancing antiviral activity is the application of a prodrug that
releases the nucleotide analog inside the cell by pH driven or enzymatic
hydrolysis. The cycloSal approach has been applied to many different
nucleosides, sometimes resulting in an impressive boost of activity.[1]
However, as the release of the nucleotide is pH driven, the prodrug is
able to enter but also to leave the cell. This is due to the fact that the
hydrolysis half-lives are typically in the range of 4 to 20 hours. A further
improvement of the cycloSal approach might therefore be the attachment
of a trigger, which is sensible to enzymes inside a cell. As esterase activity is
much higher intracellularly, we decided to use esterase cleavable moieties
that release highly polar carboxylates, trapping the prodrug inside the
cell.[2]Because first experiments with simple esters attached to the cycloSal
Address correspondence to C. Meier, University of Hamburg, Martin-Luther-King-Platz 6, Hamburg
20146, Germany. E-mail: laborhenry@gmx.de
827

828
H. J. Jessen et al.
FIGURE 1 Two different triggers (AM and POM) cleaved by esterase.
prodrug did not show enzymatic cleavage, we switched to more elaborate
groups. Acetoxymethyl- (AM), pivaloyloxymethyl-moieties (POM) were
linked to the aromatic of the cycloSal-system via a carboxylate. Indeed, these
compounds are rapidly degraded by enzymes.[3]The proposed mechanism
of degradation for these compounds is given in the Figure 1.
RESULTS
To achieve an effective “lock in,” the pH dependent half-life of the
compound should be high in contrast to a very low half-life for the en-
zyme dependent step.[4] The attachment of alkyl groups to the aromatic
of the cycloSal-systems enhances the pH dependent half-life but should
not severely influence the stability of the trigger. Therefore, a methyl
group was attached to the aromatic in position 3. The starting material 4-
hydroxy-3-methylbenzaldehyde 1 was protected with benzyl bromide at the
hydroxyl group and then subjected to a Wittig reaction in excellent yields
(Scheme 1). The newly introduced CC-double bond was hydrogenated
and the protecting group removed in a one pot reaction with Pd/C and
hydrogen. The product 2 was converted into the salicylic alcohol 3 via
hydroxymethylation in formalin and potassium hydroxide.[5]Under these
conditions, the methyl ester was also cleaved. Protection of the hydroxyl
groups as a cyclic benzylidene acetal, alkylation of the carboxylate with
pivaloyloxymethyl-iodine (POM-I), and cleavage of the protecting group
with catalytic amounts of HCl yielded the POM modified salicylic alcohol
4. Conversion into the corresponding chlorophosphite 5 was achieved by
treatment with PCl3 and pyridine.
SCHEME 1 Synthesis of chlorophosphite 5. Method a) BnBr, K2CO3 (100%); b) Wittig reagent
Ph3PCHCOOMe (92%); c) H2/Pd/C (100%), d) KOH, HCHO (100%); e) 1. Benzaldehydedimethy-
lacetale, p-TsOH (91%); 2. DIPEA, POM-I (100%); f) [HCl], reflux (78%); g) PCl3, pyridine (98%).

Intracellular Trapping of cycloSal-Pronucleotides
829
FIGURE 2 Synthesis of 3-Me-5-POM-prop-cycloSalBVDUMP 6 and –d4TMP 7.
Chlorophosphite 5 was coupled to different nucleoside analogs (BVDU,
d4T) under basic conditions and oxidized with tert-butylhydroperoxide
1
(Figure 2). H-NMR data of compounds 5, 6, and 7 are given in the ref-
erences and comments section.
Compounds 6 and 7 were hydrolyzed in an aqueous buffer (PBS, pH
7.3). The resulting products d4TMP and BVDUMP were analyzed by means
of ion-pair RP-chromatography. Another experiment was conducted in
CEM/0 (d4T) or P3HR1 (BVDU) cell extract to examine the cleavage of
the POM group. The observed hydrolysis half-lives are given in Table 1.
Finally, the stability in plasm was analyzed. In this case, no enzymatic
cleavage of the trigger was detected within a period of several hours.
CONCLUSIONS
A new and very efficient route to POM-modified cycloSal-pronucleotides
has been developed. The synthesis of the masking group proceeded in ex-
cellent yields for every single step and the obtained chlorophosphite 5 could
be coupled to different nucleoside analogs. As expected, striking differences
in hydrolysis half-lives of 6 and 7 in aqueous versus biological media were ob-
served. This behavior should eventually lead to drug enrichment inside the
cell, trapping the pronucleotide by generating a negative charge, triggered
TABLE 1 Comparison of hydrolysis half-lives
Compound t1/2 [h] (pH = 7.3)
t1/2 [h] (cell extract)
Acceleration
6
7
6.2
7.4
0.2
0.4
31
19

830
H. J. Jessen et al.
by enzymatic cleavage. Additionally, the compounds were not prone to en-
zymatic cleavage in plasm. This is the basis for an effective “lock in” mecha-
nism: prolonged hydrolysis stability in PBS (pH 7.3), reasonable stability in
plasm, and very low half-life in cell extracts, resulting in the generation of a
charged intermediate that slowly releases the drug inside the cell.
REFERENCES AND COMMENT
1. Meier, C. cycloSal-phosphates as chemical trojan horses for intracellular nucleotide and glycosyl-
monophosphate delivery — Chemistry meets biology. Eur. J. Organic Chem. 2006, 1081–1102.
2. Perigaud, C.; Gosselin, G.; Imbach, J.-L. Minireview: From the pronucleotide concept to the SATE
phosphate protecting groups. Curent Topics Medicinal Chem. 1997, 2, 15–29.
3. Pompon, A.; Lefebvre, I.; Imbach, J.-L.; Khan, S.; Farquhar, D. Decomposition pathways of the mono-
(pivaloyloxymethyl) and bis-(pivaloyloxymethyl) esters of azidothymidine-5’-monophosphate in cell
extract and in tissue culture medium — n application of the Isrp-cleaning HPLC technique, Antiviral
Chem. Chemotherapy 1994, 5, 91–98.
4. Meier, C.; Ducho, C.; Jessen, H.J.; Vukadinovic-Tenter, D.; Balzarini, J. Second-generation cycloSal-
d4TMP pronucleotides bearing esterase-cleavable sites — The “trapping” concept. Eur. J. Organic
Chem. 2005, 197–206.
5. Slough, Deutsches Patent Nr. 1931255.
1H-NMR: 5 (δ/ppm, 400 MHz, CDCl3): 6.95 (1H), 6.64 (1H), 5.73(2H),
5.39 (1H), 4.97 (1H), 2.86 (2H), 2.64 (2H), 2.20 (3H), 1.18 (9H); 6 (two
diastereoisomers, δ/ppm, 400 MHz, DMSO-d6): 11.59 (1H), 7.77 (1H), 7.28
(1H), 7.11 (1H), 6.93 (1H), 6.87 (1H), 6.15 (1H), 5.68 (2H), 5.37 (2H), 4.28
(3H), 3.93 (1H), 2.77 (2H), 2.68 (2H), 2.52 (3H), 2.17 (5H), 1.09 (9H); 7
(two diastereoisomers,δ/ppm, 400 MHz, CDCl3): 8.00 (1H), 7.22 (1H), 7.01
(1H), 6.99 (1H), 6.76 (1H), 6.36 (1H), 5.93 (1H), 5.74 (2H), 5.25 (2H), 5.01
(1H), 4.36 (2H), 2.88 (2H), 2.64 (2H), 2.23 (3H), 1.73 (3H), 1.19 (9H).