Synthesis of fluorinated acyclic nucleoside phosphonates with 5-azacytosine base moiety

Journal Pre-proof

Synthesis of fluorinated acyclic nucleoside phosphonates with 5-azacytosine base

moiety

Karel Pomeisl, Marcela Krečmerová, Radek Pohl, Robert Snoeck, Graciela Andrei

PII:

S0040-4020(19)30862-2

https://doi.org/10.1016/j.tet.2019.130529

TET 130529

DOI:

Reference:

To appear in: Tetrahedron

Received Date: 23 April 2019

Revised Date: 30 July 2019

Accepted Date: 13 August 2019

Please cite this article as: Pomeisl K, Krečmerová M, Pohl R, Snoeck R, Andrei G, Synthesis of

fluorinated acyclic nucleoside phosphonates with 5-azacytosine base moiety, Tetrahedron (2019), doi:

https://doi.org/10.1016/j.tet.2019.130529.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition

of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of

record. This version will undergo additional copyediting, typesetting and review before it is published

in its final form, but we are providing this version to give early visibility of the article. Please note that,

during the production process, errors may be discovered which could affect the content, and all legal

disclaimers that apply to the journal pertain.

Synthesis of fluorinated acyclic nucleoside

phosphonates with 5-azacytosine base moiety

Karel Pomeisl^1,2, Marcela Krečmerová^1,*, Radek Pohl¹, Robert Snoeck³and Graciela Andrei³

¹Institute of Organic Chemistry and Biochemistry, Czech Academy of Sciences, Flemingovo

2

nám. 2, CZ-166 10, Prague 6, Czech Republic; Institute of Physics, Czech Academy of

3

Sciences, Na Slovance 1999/2, 182 21 Prague 8, Czech Republic; Rega Institute for Medical

Research, KU Leuven, Herestraat 49, box 1043, B-3000 Leuven, Belgium

^*Corresponding author: Tel.: +420 220183475, e-mail: marcela@uochb.cas.cz

graciela.andrei@ kuleuven.ac.be

_________________________________________________________________________

Abstract: With respect to the strong antiviral activity of (S)-1-[3-hydroxy-2-

(phosphonomethoxy)propyl]-5-azacytosine various types of its side chain fluorinated

analogues were prepared. The title compound, (S)-1-[3-fluoro-2-(phosphonomethoxy)propyl]-

5-azacytosine (FPMP-5-azaC) was synthesised by the condensation reaction of (S)-2-

[(diisopropoxyphosphoryl)methoxy)-3-fluoropropyl p-toluenesulfonate with a sodium salt of

5-azacytosine followed by separation of appropriate N¹and O²regioisomers and ester

hydrolysis. Transformations of FPMP-5-azaC to its 5,6-dihydro-5-azacytosine counterpart,

amino acid phosphoramidate prodrugs and systems with an annelated five-membered

imidazole ring, i.e. imidazo[1,2-a][1,3,5]triazine derivatives were also carried out. 1-(2-

Phosphonomethoxy-3,3,3-trifluoropropyl)-5-azacytosine was prepared from 5-azacytosine

and trifluoromethyloxirane to form 3,3,3-trifluoro-2-hydroxypropyl-5-azacytosine which was

1

treated with diisopropyl bromomethanephosphonate followed by deprotection of esters.

Antiviral activity of all newly prepared compounds was studied. FPMP-5-azaC diisopropyl

ester inhibited the replication of herpes viruses with EC₅₀values that were about three times

higher than that of the reference anti-HCMV drug ganciclovir without displaying cytotoxicity.

_________________________________________________________________________

Keywords: Acyclic nucleoside phosphonates; 5-Azacytosine; Fluorinated nucleotides;

Prodrugs; Phosphonates

1. Introduction

The chemistry of acyclic nucleoside phosphonates (ANPs) is a key research topic of the

Institute of Organic Chemistry and Biochemistry in Prague. The importance of these

compounds lies in their biological activities, especially antiviral.^1-4Three compounds of this

class have already became commercially available drugs: cidofovir for the treatment of

human cytomegalovirus (CMV) retinitis in AIDS patients; adefovir in its prodrug form

adefovir dipivoxil for the treatment of hepatitis B (HBV); and tenofovir, either as tenofovir

disoproxil fumarate, or more recently (since 2015) also as a new prodrug form tenofovir

alafenamide (TAF), for the treatment of HIV and HBV infections.⁵The introduction of

tenofovir to the market had a crucial influence on the transformation of HIV/AIDS from a

mortal threat to a manageable chronic disease.⁶Besides, long-term systematic investigation of

ANPs has enabled the development of many other therapeutically promising structures, which

have been synthesised but have never advanced to the stage of preclinical/clinical

investigations. Among others, this concerns fluorinated purine 3-fluoro-2-

[(phosphonomethoxy)propyl] derivatives⁷and acyclic nucleoside phosphonates with 5-

azacytosine base moiety^8-10

.

2

5-Azacytosine ANPs were originally synthesized with the aim to search for new epigenetic

(hypomethylating) agents; contrary to 5-azacytosine nucleosides (azacitidine, decitabine),

however, this activity has not been found. It is evident that the presence of the sugar moiety is

essential for this effect.¹¹On the other hand, we found a new class of antiviral agents selective

against DNA viruses. The lead compound of this group is 5-azacytosine analogue of

cidofovir, 1-(S)-[3-hydroxy-2-(phosphonomethoxy)propyl]-5-azacytosine, HPMP-5-azaC

(Fig. 1). In comparison with cidofovir, its antiviral activity data are similar, but the toxicity

profile is more favourable, with the selectivity index (the ratio of the 50% cytotoxic

concentration (CC₅₀) to the 50% effective concentration (EC₅₀)) being 2- to 16-fold higher

than that of cidofovir.⁸The antiviral activity of HPMP-5-azaC has been increased many times

by the transformation to ester prodrugs derived from its cyclic form.⁹The only disadvantage

of HPMP-5-azaC is its rather complicated metabolic profile caused by its chemical (and also

enzymatic) instability.^12,13Therefore, the compound was finally not advanced to the stage of

clinical investigations. Other 5-azacytosine ANPs have better stability but they are either

inactive or their antiviral activity is only marginal. Interestingly, some of them, e.g. 2-

(phosphonomethoxy)ethyl derivative exhibited antiviral activity when converted into

appropriate prodrugs.¹⁰

Figure 1. The structures of anti-DNA virus agent cidofovir and its 5-azacytosine analogue,

HPMP-5-azaC.

3

In this study, we investigate a new type of 5-azacytosine ANPs modified with fluorine at

the acyclic part. Fluorinated derivatives, especially nucleoside and nucleotide analogues are

compounds of great interest.¹⁴An introduction of the fluorine atom into biologically active

molecules often leads to remarkable changes in their physical, biological, and

pharmacokinetic properties due to its electronegativity, size, omniphobicity or lipophilicity

and electrostatic interactions.¹⁵In clinical practice, fluorinated nucleosides are utilized as

effective cytostatics (e.g. gemcitabine) or antivirals (trifluridine, sofosbuvir).^16,17Within the

group of ANPs, 3-fluoro-2-(phosphonomethoxy)propyl (FPMP) derivatives have been

identified as potent and selective inhibitors of retroviruses. The adenine derivative, (S)-9-[3-

fluoro-2-(phosphonomethoxy)propyl] adenine, ((S)-FPMPA) exhibits even better in vivo anti-

HIV and anti-HBV parameters than adefovir. The activity against HIV is also known for both

enantiomers of 9-[3-fluoro-2-(phosphonomethoxy)propyl]-2,6-diaminopurine (FPMPDAP)

and its guanine counterpart.^7,18Structures of these compounds are depicted in Fig. 2.

Figure 2. The structures of the antiretroviral 3-fluoro-2-(phosphonomethoxy)propyl

derivatives of purine bases.

This paper describes the synthesis and biological activities of side-chain fluorinated analogues

of HPMP-5-azaC and their derivatives with further modifications on the base moiety. These

modifications have been designed with the aim to stabilise the 5-azacytosine ring.

4

Moreover, it must be taken into account that ANPs as compounds bearing (a) free

phosphonate group(s) have a polar character responsible for their limited intracellular delivery

and low oral bioavailability. The usual way to overcome this problem is the transformation of

such compounds to appropriate prodrugs. The most frequently used ANP prodrugs are

acyloxyalkyl (pivaloyloxymethyl, POM), alkyloxyalkyl, S-acylthioethyl (SATE), aryl, cyclic

1-aryl-1,3-propanyl and cyclosaligenyl phophonate esters, phosphoramidates bearing amino

acid esters and proTides.¹⁹Despite promising in vitro or in vivo studies, only a few of them

have been proceeded into clinical trials. This can be attributed to the difficulty of the

achievement of the right prodrug property balance, in particular suitable chemical stability,

rate of conversion, safety with respect to by-products, the avoidance of stereoisomers,

adequate physical properties (the ability to crystallise, solubility), and the ability to synthesize

substantial quantities. The synthesis of ANP prodrugs and the investigation of their

pharmacological properties have been the subject of many reviews.^20,21

Regarding the above-mentioned facts, the transformation of our new structures to prodrug

forms is an integral part of this work.

2. Chemistry

The most logical method of the preparation of FPMP-5-azaC reveals direct

condensation of the sodium salt of 5-azacytosine with an appropriate fluorinated

precursor. Such a reactive precursor can be (S)-2-[(diisopropoxyphosphoryl)methoxy)-

3-fluoropropyl p-toluenesulphonate (FPMP synthon, 1) or its methanesulphonyl

analogue. The synthon approach has already been reported for the preparation of purine

FPMP derivatives by Jindrich et al.²²Their synthesis started from protected 5´-O-

ribonolactone and included as many as nine reaction steps. Another, more efficient

synthesis of FPMP synthons utilises enantiomeric fluorohydrines accessible from

5

commercially available O-tritylated glycidols by reaction with potassium hydrogen

difluoride.²³In our present work, we have investigated the fluorination of (R)-3-

benzyloxy-2-[(diisopropylphosphoryl)methoxy)]propanol (3), a compound utilised in

our laboratory as a common starting material for the preparation of 3-hydroxy-2-

(phosphonomethoxy)propyl derivatives. This synthon is accessible on a large scale by a

three-step synthesis from (S)-2-(trityloxymethyl)oxirane (2) (Scheme 1).²⁴

Scheme 1. The synthesis of (S)-2-[(diisopropoxyphosphoryl)methoxy)-3-fluoropropyl p-

toluenesulphonate (FPMP synthon, 1)

Our first experiments with the fluorination of 3 were carried out via previous activation

of the hydroxyl group, e.g. by the reaction with p-toluenesulphonyl chloride.²⁵Although the

reaction of tosylate 4, formed in this way, with tetrabutylammonium fluoride proceeded

satisfactorily, the total preparative yield of the fluoro derivative 5 was not high because of

the complicated and laborious isolation and purification of the tosylate intermediate.

An alternative way is the direct fluorination of the hydroxy derivative 3. This approach

has never been explored in the area of ANP synthons whose synthesis is usually performed

as a large-scale multi-step process without the isolation of intermediates. The fluorination

6

reagents generally used for the substitution of hydroxy groups are DAST,²⁶Olah’s

reagent,²⁷and fluoroalkylamine reagents.²⁸

We have recently described perfluorobutane-1-sulphonyl fluoride as a nucleophilic

fluorination reagent for the preparation of thymidine phosphorylase inhibitors (S) and (R)-1-

[3-fluoro-2-(phosphonomethoxy)propyl]thymines (FPMPT).^29-31Unlike various strongly

corrosive reagents, e.g. DAST or deoxofluor, perfluorobutane-1-sulphonyl fluoride is

stable, only little moisture-sensitive and easy to handle. Its use for the direct fluorination of

ANPs is known only from for several examples. In the above mentioned FPMPT synthesis,

the appropriate starting material was the 3-hydroxy-2-[(phosphonomethoxy)propyl]

derivative of 4-O-methoxy 5-methylpyrimidinone (not thymine). The reason is that these

fluorinations are sensitive to the presence of acidic hydrogen atoms in the starting molecule.

For example, the performance of the same reaction with the diisopropyl ester of HPMPT

(containing NH at the position 3 of the thymine moiety), results in elimination under basic

conditions resulting in a six-membered ring closure through the side-chain hydroxy group.²⁹

Similarly, this fluorination method also fails with cytosine derivatives (HPMPC).

An important aspect of syntheses with perfluorobutane-1-sulphonyl fluoride is the

selection of reaction conditions: base and solvent.³²The base should be a good proton

acceptor with low nucleophilicity and moderate basicity to avoid elimination. It the

preparation of fluoro derivative 5 from (R)-3-benzyloxy-2-

[(diisopropylphosphoryl)methoxy)]propanol (3), we chose between N-ethylpiperidine

(NEP) and 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU). The reaction with NEP had been

reported to provide better results in the syntheses of fluorinated nucleosides,³²but the

preparation of our FPMP precursor proceeded satisfactory using DBU. The reaction was

carried out in toluene, initially at room temperature. Although the yields of 5 were 42–50%

the reaction rate was slow. A significant decrease of reaction time was achieved by heating

7

to 90 °C. Full conversion was observed when DBU was added for a few hours (3-4 h). The

course of fluorination reaction with NEP was worse and required excess of

perfluorobutanesulphonyl fluoride (90% conversion after 9 h). The final FPMP synthon,

(S)-2-[(diisopropoxyphosphoryl)methoxy)-3-fluoropropyl p-toluenesulphonate (1), was

obtained from 5 by the removal of a benzyl group via catalytic hydrogenation to give

intermediate 6 which was further subjected to tosylation (Scheme 1).

The condensation of (S)-2-[(diisopropoxyphosphoryl)methoxy)-3-fluoropropyl p-

toluenesulphonate (1) with a sodium salt of 5-azacytosine (7) was performed by heating in

N,N-dimethylformamide with catalytic amount of caesium carbonate. Reaction afforded a

mixture of N¹and O²regioisomers 8a and 8b, whose ratio depended on the reaction time

(Scheme 2). The highest regioselectivity was observed when 5-azacytosine sodium salt was

heated with phosphonate 1 at 120 °C for 11 h. In this case, only N¹-isomer 8a was detected

but its preparative yield was low (28 %). Further extension of the reaction time resulted in

increased formation of O²-isomer 8b. After 24 h at 120 °C, the ratio of regioisomers 8a:8b

was 7:4 while their preparative yields were 28 % (8a) and 16 % (8b). Regarding these

observations, it is obvious that the alkylation is a balanced process and the desired N¹-

isomer is formed as a product possessing kinetic stability. The final product, free

phosphonic acid 9 (FPMP-5-azaC) was obtained from diisopropyl ester 8a by the standard

procedure of deprotection by bromotrimethylsilane followed by hydrolysis.

8

Scheme 2. The condensation of (S)-2-[(diisopropoxyphosphoryl)methoxy)-3-fluoropropyl

p-toluenesulphonate (FPMP synthon, 1) with a sodium salt of 5-azacytosine (7)

In order to have also its prodrug form available for antiviral screening we selected its

phosphoramidate prodrug 11. The synthesis was performed by the modified procedure

originally developed by Jansa³³based on the reaction of diisopropyl ester 8a with

bromotrimethylsilane, first giving the intermediary bis(trimethylsilyl) ester 10. This

intermediate can be converted without isolation to the desired bis(amidate) 11 by a reaction

with an appropriate amino acid ester activated by 2,2´-dithiopyridine (Aldrithiol^TM) and

triphenylphosphine (Scheme 3). A direct reaction of 9 with ethyl L-alanine, Aldrithiol^TM

and triphenylphosphine was not successful, giving the desired bis(amidate) 11 in low yields

only, together with a number of by-products.

9

Scheme 3. The synthesis of the bis(amide) prodrug of 1-(S)-[3-fluoro-2-

(phosphonomethoxy)propyl]-5azacytosine

A significant problem of 5-azacytosine compounds is their hydrolytic instability which

substantially compromise their pharmaceutical potential.^11-13This instability is caused by the

electron deficiency in position 6 of the triazine ring, where the imine-resembling carbon could

be easily attacked by a nucleophile, e.g. an hydroxyl ion, accompanied by the cleavage of the

ring.^11,12On the other hand, the conversion of 5-azacytosine compounds into their 5,6-

dihydro-5-azacytosine counterparts with sp³-hybridisied carbon (CH₂) at the position 6, leads

to the compounds which are hydrolytically stable. Nucleosides modified in this way, e.g. 5,6-

dihydro-5-azacytidine and 2´-deoxy-5,6-dihydro-5-azacytidine, still preserve their antitumor

and hypomethylating activities. They are lower in comparison with the original 5-azacytosine

derivatives but the compounds are less toxic.^11,34In the field of ANPs, we previously reported

the 5,6-dihydro derivative of HPMP-5-azaC (Figure 1). The compound has preserved its anti-

DNA virus activity (albeit lower than HPMP-5-azaC); nevertheless in contrast to HPMP-5-

azaC, it also exhibits inhibitory activity against HCV replication, and is hydrolytically stable.⁸

The transformation of FPMP-5-azaC into the corresponding 5,6-dihydro derivative 13 was

carried out on the level of its diisopropyl ester 8a via catalytic hydrogenation. This step was

followed by the deprotection of ester groups in the intermediate 12 using

bromotrimethylsilane (Scheme 4).

10

Scheme 4. The synthesis of 1-(S)-5,6-dihydro-[3-fluoro-2-(phosphonomethoxy)propyl]-5-

azacytosine (13).

The guanidine grouping of the 5,6-dihydro-5-azacytosine moiety provides a unique

opportunity for the preparation of imidazo[1,2-a][1,3,5]triazine derivatives – compounds with

four hydrogen-bonding sites bearing an annelated 5-membered imidazole ring. These

analogues can be formed by reaction of 5,6-dihydro-5-azacytosine derivatives with

appropriate carbonyl compounds. We have described the coupling of 5,6-dihydro-5-

azacytosine phoshonate 12 with various 2-bromoacetophenones 14a-c. (Scheme 5). Although

the desired annelated products 15a-c were formed in all cases, the imidazole imino group was

found sensitive towards further alkylation with some 2-bromoacetophenones. When the

compound 12 was treated with 2-bromoacetophenones 14b and 14c alkylated by-products 16b

and 16c were isolated together with products 15b, and 15c. On the other hand, 2-

bromoacetophenone 14a did not afford any further alkylated product.

11

Scheme 5. The transformation of 5,6-dihydro-5-azacytosine ANPs to their imidazo[1,2-

a][1,3,5]triazine derivatives 15a-c.

Surprisingly enough, a serious problem occurred in the deprotection of diisopropyl ester

groups to the final phosphonic acids 17a-c. The usual way using bromotrimethylsilane

completely failed, leading to a mixture of decomposition products. Another possibility was a

microwave-assisted acid hydrolysis. This method of the transformation of phosphonate esters

to free ANPs has been originally developed at the IOCB by Jansa and Holý. This method is

usually very clean and the isolation of products is easy.^35,36In our experiments, we

hydrolysed 15a-c with two equivalents of HCl at 150 °C for 5-15 min under micro-wave

irradiation. Unfortunately, the ring of the triazine moiety opened and the C-6 methylene group

was hydrolytically split off. This indicated the destabilisation of the 5,6-dihydro-5-

azacytosine ring as a consequence of imidazole-triazine annelation. On the other hand, we had

observed earlier that 5-azacytosine ANPs can have better stability in the cases when their

phosphonic function is masked by an appropriate prodrug group.^9,11,37Taking into account

12

this finding, we selected another synthetic strategy starting from bis(amidate) prodrug of

FPMP-5-azaC 11. Compound 11 can be easily reduced to the corresponding 5,6-dihydro form

18 which is then subjected to condensation reaction with 2-bromoacetophenones to form the

desired annelated structures 19a-c in their prodrug forms (Scheme 6). The other way, Jansa´s

method³³for the transformation of phosphonate ester 12 into phosphoramidate 18 doesn´t take

place. This may have been caused by the presence of acetic acid in the structure of 12.

Condensation reactions using 2,2’-dipyridyldisulphide (Aldrithiol) and triphenylphosphine are

known to proceed via a cyclic intermediate whose formation requires a proton coming from a

nucleophile³⁸(e.g. amino acid ester). It is possible that acetic acid preferentially protonates a

pyridine unit of 2-thiopyridine which then cannot activate a nucleophile and phosphonate

units.

Scheme 6. The synthesis of the bis(amidate) prodrug of 5,6-dihydro-FPMP-5azaC (18) and its

imidazo[1,2-a][1,3,5]triazine analogues (19a-c).

Besides FPMP derivatives, we also worked on the syntheses of 5-azacytosine ANPs

bearing gem-difluorinated arrangement by the method developed in our laboratory for purine

13

gem-difluorinated ANPs.³⁶Unfortunately, our effort failed because of the instability of the

triazine ring during the deprotection of phosphonate ester groups (with bromotrimethylsilane

as well as under microwave conditions).

On the other hand, ANPs with trifluoromethyl arrangement are known to be stable towards

Lewis acids. Therefore, 5-azacytosine phosphonate 25 could be prepared quite easily (Scheme

7). To a first approximation, the compounds were prepared as racemates using cheap and

easily accessible trifluoromethyloxirane. This approach is sufficient for the evaluation of the

true biological potential of the compounds. If any biological activity were sought, pure

enantiomers would have to be subsequently synthesised.

The synthesis is based on the treatment of 5-azacytosine (20) with trifluoromethyloxirane

(21) followed by the reaction of the formed hydroxy derivative 22 with appropriate

phosphonate synthon to give the desired phosphonate in the form of ester 24.

For this purpose, two phosphonate synthons were investigated. While the reaction with

diisopropyl tosyloxymethanephosphonate and sodium hydride afforded only low yields of 24,

the conversion increased substantially when more reactive diisopropyl

bromomethanephosphonate and sodium tert-butanoate were used. In agreement with our

previous observations, the presence of fluorine atom or trifluoromethyl arrangement in the

intermediate (e.g. 22) may significantly decrease the nucleophilicity of the secondary alcohol

group compared to corresponding non-fluorinated analogues.³⁶The final deprotection of ester

groups with bromotrimethylsilane afforded free phosphonic acid 25. Additionally, the

catalytic hydrogenation of 5-azacytosine derivative 22 in the presence of acetic acid afforded

5,6-dihydro derivative 23 in the form of an acetate salt.

14

Scheme 7. The synthesis of 5-azacytosine ANPs with trifluoromethyl arrangement

3. Biological activity

The synthesised side-chain fluorinated analogues of HPMP-5-azaC were evaluated for their

inhibitory activity against a wide range of DNA and RNA viruses: in human embryonic lung

(HEL) cells [herpes simplex virus-1 (KOS strain), herpes simplex virus-2 (G strain),

thymidine kinase deficient (acyclovir resistant) herpes simplex virus-1 (TK– KOS ACVr),

vaccinia virus, vesicular stomatitis virus, human cytomegalovirus (HCMV) (AD-169 strain

and Davis strains), varicella-zoster virus (VZV) (TK+ VZV strain and TK– VZV strains))], in

HeLa cell cultures [vesicular stomatitis virus, Coxsackie virus B4 and respiratory syncytial

virus], in Vero cell cultures [para-influenza-3 virus, reovirus-1, Sindbis virus, Coxsackie virus

B4, Punta Toro virus, yellow fever virus], in CRFK cell cultures [feline corona virus (FIPV)],

and in MDCK cell cultures [influenza A virus (H1N1 and H3N2 subtypes) and influenza B

virus]. Ganciclovir, cidofovir, acyclovir, brivudin, zanamivir, amantadine, rimantadine,

ribavirin, dextran sulphate (molecular weight 10000, DS-10000), mycophenolic acid,

15

Hippeastrum hybrid agglutinin (HHA), and Urtica dioica agglutinin (UDA) were used as the

reference compounds. The antiviral activity was expressed as EC₅₀, i.e. the compound

concentration required to decrease virus plaque formation (VZV) or virus-induced

cytopathogenicity (other viruses) by 50%. While none of the compounds showed any activity

against RNA viruses, compounds 8a and 22 were able to inhibit, albeit weakly, the replication

of herpesviruses (Table 1). The FPMP-5-azaC diisopropyl ester (8a) inhibited the replication

of HCMV with EC₅₀values that were about three times higher than those of the reference

anti-HCMV drug ganciclovir without displaying cytotoxicity for HEL cells up to the highest

concentration tested (100 µM). The compound 8a proved significantly less active against

wild-type strains of HSV-1, HSV-2 and VZV than the reference drug acyclovir but was in

contrast to acyclovir remained active against TK-deficient viruses. This derivative proved

inactive against other DNA viruses such as vaccinia virus (poxvirus) and adenoviruses.

Compound 22 (3,3,3-trifluoro-2-hydroxypropyl-5-azaC) showed marginal activity against

herpesviruses and adenovirus 2 with EC₅₀values in the range of 40-270 µM.

Table 1. The in vitro antiviral activity of fluorinated 5-azacytosine derivatives against DNA

viruses compared to reference antiviral drugs acyclovir (ACV), brivudine (BVDU),

ganciclovir (GCV) and cidofovir (CDV).

Cytotoxicity

Antiviral activity - EC₅₀(µM)^a

(µM)

HCMV

VZV

HSV-1

HSV-2 G

Kos

HSV-1

c

MCC^bCC₅₀

No.

VACV Adeno-2

(TK⁺)

ACV-R (TK^-)

AD-169

Davis

Oka (TK⁺)

07/1 (TK^-)

(TK⁺)

24.5 ±

13.6

7.9 ±

5.5

11.7 ±

4.0

8a

8.8 ± 0.2

39 ± 19

17.5 ± 10.6

32.5 ± 12.5

>100

9

>100

ND^d

ND

11

13

18

19a

16

19b

19c

>100

ND

231 ±

12.9

200 ±

17.8

40.6 ±

11.7

22

126 ± 13.8

230 ± 29

>446

268 ± 178

>446

23

25

>100

ND

1.44 ±

0.79

0.48 ±

0.27

ACV

ND

43.9 ± 19.8

26.7 ± 19.0

ND

0.21 ± 0.12

123 ± 26

0.03 ± 0

48.9 ± 57.0

1.6 ± 0.6

>250

ND

>440

>300

>350

>300

>440

>300

>350

>300

0.023 ±

0.021

0.03 ±

0.02

11.3 ±

6.4

BVDU ND

3.0 ±

2.7

0.03 ±

0.01

GCV

CDV

6.7 ± 2.9

ND

>100

1.1 ± 0.7

1.3 ± 0.3

0.93 ±

0.34

0.73 ±

0.31

ND

1.9 ± 1.2

1.3 ± 0.4

26 ± 12 8.3 ± 2.1

^aThe effective concentration required to reduce virus plaque formation (VZV) or virus-induced cytopathic effect

(HCMV, HSV-1, HSV-2, vaccinia virus (VACV) and adenovirus-2) by 50%. The virus input was (1 CCID₅₀

being the virus dose to infect 50% of the cell cultures) or 20 plaque forming units (PFU) in the case of VZV.

^bThe minimum cytotoxic concentration that causes a microscopically detectable alteration of cell morphology.

^cThe cytostatic concentration required to reduce cell growth by 50%.

^dNot determined

4. Conclusions

Several synthetic strategies for side-chain fluorinated acyclic nucleotide analogues bearing 5-

azacytosine and imidazo[1,2-a][1,3,5]triazine base moieties have been developed. Highly polar

free phosphonic acids were transformed into amidate prodrugs. All new compounds were

subjected to the investigation of biological (antiviral) activities. Except for the FPMP 5-azaC

diisopropyl ester (8a), which had some activity against human cytomegalovirus and

herpesviruses and for 1-(3,3,3-trifluoro-2-hydroxypropyl)-5-azacytosine (22) with negligible

activity against herpesviruses and adenovirus, none of the newly synthesized acyclic

nucleoside analogues (ANPs) exhibited antiviral activity.

5. Experimental

5.1. General

17

Unless stated otherwise, solvents were evaporated at 40 °C/2 kPa and compounds were dried

at 13 Pa. Column chromatography was performed on silica gel 60 ꢀm (Fluka). Analytical

TLC was performed on silica gel 60 F₂₅₄plates (Merck) in the solvent systems: chloroform-

methanol 7:1 (S1), chloroform-methanol 7:2 (S2) and 10:1 (S3), ethyl acetate-acetone-

ethanol-water 18:3:2:2 (S4) or 16:3:3:2 (S5), 2-propanol-25% aq. ammonia-water 7:1:2 (S6).

Preparative TLC was carried out on 45 x 18 x 0.4 cm loose-layer silica gel containing UV

indicator (S7). Compound 13 was purified on a column (3.3 x 1.7 cm) of spherical reverse

phase silica gel C18 (45–75 µm), 70 Å (Versa Flash, Supelco), (S8). Microwave-assisted

hydrolysis was performed on a CEM Discover reactor (USA) under the following conditions:

the power of 50-150 W, the maximum pressure of 1.7 MPa (17 Bar), the temperature profile

of 0˗150 °C (1 min), followed by 150 °C (25 min). The reactions were performed in a high-

pressure test tube (10 mL), equipped with a septum and a magnetic stirrer. Reaction time was

modified according to the rate of conversion of the starting esters. IR spectra were recorded

on a FTIR Spectrometer Bruker IFS 55 (Equinox) in KBr, CCl₄and CHCl₃. ¹H and ¹³C NMR

spectra were acquired on Bruker Avance 600 III HD and Bruker Avance 500 III HD

spectrometers operating at 600.1 and 500.0 MHz for ¹H and 150.9 and 125.7 MHz for ¹³C,

respectively. ³¹P{¹H} NMR spectra were recorded on Bruker Avance 500 III HD or Bruker

Avance 400 III HD operating at 202.3 or 162.0 MHz, respectively. ¹⁹F NMR spectra were

recorded on Bruker Avance 500 III HD operating at 470.3 MHz. NMR samples were

dissolved in CDCl₃, DMSO-d₆, CD₃OD or D₂O and NMR spectra were referenced to the

residual solvent signal (CDCl₃: δ(¹H) = 7.26 ppm, δ(¹³C) = 77.0 ppm; DMSO-d₆: δ(¹H) = 2.50

ppm, δ(¹³C) = 39.7 ppm; CD₃OD-d₆: δ(¹H) = 3.31 ppm, δ(¹³C) = 49.0 ppm). Aqueous samples

were referenced to signals of 1,4-dioxane as an external standard (δ(¹H) = 3.75 ppm, δ(¹³C) =

69.3 ppm). The complete assignment of NMR signals was based on a combination of 1D and

2D correlation experiments (H,H-COSY, H,C-HSQC, H,C-HMBC). The general numbering

18

scheme for the assignment of NMR signals is outlined in Fig. 3. Mass spectra were measured

on an LTQ Orbitrap XL (Thermo Fisher Scientific) operated in the ESI mode. Optical

rotations were measured on a Autopol IV polarimeter (Rudolph Research Analytical, U.S.A.)

at 20 °C, [α]_Dvalues are given in 10^-1deg cm²g^-1. The concentrations for specific rotation

measuements are in the units of g/100 mL.

5.2. Materials and Chemicals

Standard chemicals and ion-exchange resins were purchased from Sigma-Aldrich (Czech

Republic). The synthetic precursor 3 was prepared from tritylglycidyl ether (2) according to

ref.²⁴The sodium salt of 5-azacytosine 7 was prepared by the treatment of 5-azacytosine with

sodium methoxide according to ref. ²⁴Diisopropyl bromomethylphosphonate was purchased

from Alfa Aesar. Trifluoromethyloxirane was purchased from Fluorochem (Wesley Street,

Old Glossop). Anhydrous solvents were purchased from Sigma-Aldrich and Penta (Czech

Republic).

Figure 3. The general numbering scheme for assignment of NMR signals.

5.3. (S)-3-(Benzyloxy)-2-[(diisopropoxyphosphoryl)methoxy]propyl 4-

methylbenzenesulphonate (4).

A solution of 3 (63 g, 138 mmol) in acetonitrile (400 mL) was cooled to 0 °C and

triethylamine (159 mL, 1.1 mol), TsCl (40 g, 166 mmol) and DMAP (50 mg) were added. The

19

mixture was stirred at room temperature for 54 h (until full conversion), then concentrated in

vacuo and the residue was partitioned between ether and water (600 mL/100 mL). An organic

layer was dried over magnesium sulphate and evaporated. The obtained material was used for

the next step without purification. Yield: 51 g (58 %) of 4 as yellow oil. ¹H, and ¹³C NMR

data were identical to those reported in ref.³⁹

5.4. Diisopropyl (S)-{[(1-(benzyloxy)-3-fluoropropan-2-yl)oxy]methyl}phosphonate (5).

5.4.1. Method 1. A mixture of 4 (51 g, 99.1 mmol) and 1M TBAF solution in THF (147 mL)

was refluxed vigorously in dioxane (500 mL) at 100-110 °C for 6.5 h. The mixture was

concentrated in vacuo and the residue was partitioned between ether and water (500 mL/60

mL). An aqueous layer was extracted with ether (2x 200 mL). Combined organic extracts

were dried over MgSO₄and evaporated. The residue was used without further purification.

Yield: 31 g (86 %) of 5 as yellow oil; [α]_D+13.6 (c 0.295, CHCl₃). IR: (CCl₄) ν_max3090,

3066, 3033, 2981, 2934, 1635, 1604, 1586, 1495, 1453, 1467, 1386, 1376, 1315, 1267, 1191,

1178, 1142, 1071, 1106, 1030, 1010, 991, 889, 710, 611 cm^-1. ESIMS m/z 363.1 (M + 1)⁺

(48). HRMS m/z (ESI) calcd. for C₁₇H₂₈O₅FNaP (M + 1)⁺385.1551 found 385.1551. ¹H NMR

(500.0 MHz, CDCl₃): 1.308, 1.314, 1.321, 1.328 (4 × d, 4 × 3H, J_vic= 6.2, (CH₃)₂CH); 3.58

(ddd, 1H, J_gem= 10.2, J_1ʹb,2ʹ= 5.5, J_H,F= 1.2, H-1ʹb); 3.62 (ddd, 1H, J_gem= 10.2, J_1ʹa,2ʹ= 5.4,

J_H,F= 1.9, H-1ʹa); 3.89 (m, 1H, H-2ʹ); 3.93 (d, 2H, J_H,P= 8.7, H-4ʹ); 4.51 (ddd, 1H, J_H,F

=

47.3, J_gem= 10.1, J_3ʹb,2ʹ= 5.6, H-3ʹb); 4.53 (s, 2H, CH₂Ph); 4.57 (ddd, 1H, J_H,F= 47.3, J_gem

=

10.1, J_3ʹa,2ʹ= 3.6, H-3ʹa); 4.68 – 4.80 (m, 2H, CH(CH₃)₂); 7.25 – 7.37 (m, 5H, H-o,m,p-Ph).

¹³C NMR (125.7 MHz, CDCl₃): 23.9 (d, J_C,P= 4.6, (CH₃)₂CH); 24.1 (d, J_C,P= 3.8,

(CH₃)₂CH); 65.2 (d, J_C,P= 167.9, CH₂-4ʹ); 68.5 (d, J_C,P= 7.8, CH₂-1ʹ); 71.2, 71.2 (2 × d, J_C,P

= 6.5, CH(CH₃)₂); 73.5 (CH₂Ph); 79.3 (dd, J_C,F= 18.8, J_C,P= 10.7, CH-2ʹ); 83.2 (d, J_C,F

171.2, CH₂-3ʹ); 127.6 (CH-o-Ph); 127.8 (CH-p-Ph); 128.4 (CH-m-Ph); 137.7 (C-i-Ph).

=

20

³¹P{¹H} NMR (202.3 MHz, CDCl₃): 19.41. ¹⁹F NMR (470.4 MHz, CDCl₃): -226.78 (td, J =

47.3, 19.1).

5.4.2. Method 2. A mixture of DBU (250 ꢀL, 1.7 mmol) or N-ethylpyrrolidine (86 ꢀL,0.6

mmol) and 92% perfluorobutane-1-sulphonyl fluoride (410 mg, 1.2 mmol) was stirred in

toluene (1 ml) at room temperature for 10 min. A solution of 3 (300 mg, 0.8 mmol) in toluene

(3 ml) was added and the resulting mixture was heated at 90 °C for 4 h (TLC control in ethyl

acetate). The mixture was evaporated and the residue was chromatographed on a silica gel

column in a gradient of hexane (100%) to hexane – ethyl acetate (2:9). The yield of 5: 150 mg

(50 %, reaction with DBU) or 126 mg (42 %, reaction with NEP), yellow oil. ¹H, and ¹³C

NMR data were identical to those described in Method 1.

5.5. Diisopropyl (S) {[(1-fluoro-3-hydroxypropan-2-yl)oxy]methyl}phosphonate (6).

A solution of 5 (31 g, 85.6 mmol) in methanol (300 mL) was hydrogenated in a high pressure

autoclave (10 atm) over 10% palladium on charcoal (20 g, suspended in water (20 mL)) for 48

h. The mixture was filtered through a Celite pad and the filtrate was concentrated in vacuo.

The residue was chromatographed on silica gel in ethyl acetate. Yield: 13.3 g (57 %) of 6 as

yellowish oil. ESIMS m/z 273.1 (M + 1)⁺(22). HRMS m/z (ESI) calcd. for C₁₀H₂₃O₅FP (M +

1)⁺273.1261, found 273.1262. ¹H NMR (500.0 MHz, CDCl₃): 1.33, 1.346, 1.349 (3 × d, 12H,

J_vic= 6.2, (CH₃)₂CH); 3.10 (bs, 1H, OH); 3.64 (m, 1H, H-1ʹb); 3.74 – 3.81 (m, 2H, H-1ʹa,2ʹ);

3.81 (dd, 1H, J_gem= 14.1, J_H,P= 8.4, H-4ʹb);4.05 (dd, 1H, J_gem= 14.1, J_H,P= 7.5, H-4ʹa); 4.40

– 4.56 (m, 2H, H-3ʹ); 4.69 – 4.83 (m, 2H, CH(CH₃)₂). ¹³C NMR (125.7 MHz, CDCl₃): 23.9

(d, J_C,P= 4.6, (CH₃)₂CH); 24.0 (d, J_C,P= 4.2, (CH₃)₂CH); 24.1 (d, J_C,P= 3.8, (CH₃)₂CH); 61.0

(d, J_C,P= 7.9, CH₂-1ʹ); 65.4 (d, J_C,P= 169.0, CH₂-4ʹ); 71.4 (d, J_C,P= 6.9, CH(CH₃)₂); 71.8 (d,

J_C,P= 6.6, CH(CH₃)₂); 82.2 (dd, J_C,F= 18.7, J_C,P= 7.9, CH-2ʹ); 82.9 (d, J_C,F= 170.8, CH₂-3ʹ).

21

³¹P{¹H} NMR (202.3 MHz, CDCl₃): 20.63. ¹⁹F NMR (470.4 MHz, CDCl₃): -226.42 (td, J =

47.2, 17.0).

5.6. (S)-2-[(Diisopropoxyphosphoryl)methoxy]-3-fluoropropyl 4-

methylbenzenesulphonate (1).

Triethylamine (18 mL, 129 mmol), tosyl chloride (7.31 g, 38.3 mmol) and

dimethylaminopyridine (20 mg, catalytic amount) were added to a 0°C cold solution of 6 (5 g,

20.0 mmol) in acetonitrile (250 mL). The mixture was stirred at room temperature for 48 h,

then evaporated, and the residue was extracted with a mixture of ether and brine (200 mL/50

mL). The organic layer was dried (MgSO₄) and evaporated and the residue chromatographed

on silica gel in ethyl acetate. Yield 5.1 g (65 %) of 1 as yellow oil. R_f(EtOAc) 0.50; [α]_D

+11.8 (c 0.651, CHCl₃). IR: (CCl₄) ν_max3068, 3034, 2981, 2934, 2875, 1599, 1496, 1467,

1453, 1403, 1385, 1376, 1375, 1307, 1289, 1260, 1212, 1191, 1180, 1142, 1119, 1106, 1113,

1099, 1042, 889, 833, 705, 666, 571, 555 cm^-1. ESIMS m/z 427.1 (M + 1)⁺(42). HRMS m/z

(ESI) calcd. for C₁₇H₂₉O₇FPS (M + 1)⁺427.1350, found 427.1351. Anal. Calcd. for

C₁₇H₂₈FO₇PS: C, 47.9; H, 6.6; P, 7.3; S, 7.5; F, 4.5. Found: C, 48.2; H, 6.6; P, 7.5; S, 7.2; F,

4.3. NMR data are in agreement with the literature^22,23

.

5.7. The synthesis of 5-azacytosine FPMP-derivatives 8a and 8b.

A mixture of 1 (1.5 g, 3.5 mmol), the sodium salt of 5-azacytosine²⁴(7) (456 mg, 3.4 mmol)

and caesium carbonate (10 mg, 0.03 mmol) in dimethylformamide (11 mL) was heated at 120

°C for 24 h. The mixture was concentrated in vacuo, the residue was coevaporated with

toluene (2 x 2 mL) and chromatographed on a preparative TLC plate (S7) in dichloromethane-

methanol (7:1).

22

5.7.1. Diisopropyl (S)-{[(1-(4-amino-2-oxo-1,3,5-triazin-1(2H)-yl)-3-fluoropropan-2-

yl)oxy]methyl}phosphonate (8a).

Yield: 359 mg (29 %) of a white amorphous solid. R_f= 0.40 (S1). [α]_D-55.8 (c 0.276,

CHCl₃). IR: (CCl₄) ν_max3540, 3487, 3422, 3305, 3305, 3185, 3120, 2985, 2937, 1652, 1634,

1691, 1599, 1570, 1507, 1463, 1440, 1388, 1376, 1244, 1178, 1142, 1104, 1127, 1015, 997,

890 cm^-1. ESIMS m/z 389.1 (M + 1)⁺(100). HRMS m/z (ESI) calcd. for C₁₃H₂₄O₅N₄FNaP (M

+ Na)⁺389.1361, found 389.1466. ¹H NMR (500.0 MHz, CDCl₃): 1.30, 1.31, 1.33 (3 × d,

12H, J_vic= 6.0, (CH₃)₂CH); 3.68 (dd, 1H, J_gem= 14.0, J_1ʹb,2ʹ= 8.1, H-1ʹb); 3.74 (dd, 1H, J_gem

13.7, J_H,P= 9.3, H-4ʹb); 3.93 (dd, 1H, J_gem= 13.7, J_H,P= 8.5, H-4ʹa); 4.00 (ddddd, 1H, J_H,F

=

22.8, J_2ʹ,1ʹ= 8.1, 3.1, J_2ʹ,3ʹ= 4.5, 2.7, H-2ʹ); 4.21 (dd, 1H, J_gem= 14.0, J_1ʹa,2ʹ= 3.1, H-1ʹa); 4.46

(ddd, 1H, J_H,F= 47.1, J_gem= 10.7, J_3ʹb,2ʹ= 4.5, H-3ʹb); 4.69 (ddd, 1H, J_H,F= 47.1, J_gem= 10.7,

J_3ʹa,2ʹ= 2.7, H-3ʹa); 4.69 – 4.78 (m, 2H, CH(CH₃)₂); 5.67 (bs, 1H, NH_aH_b); 6.42 (bs, 1H,

NH_aH_b); 8.04 (s, 1H, H-6). ¹³C NMR (125.7 MHz, CDCl₃): 23.9 (d, J_C,P= 4.6, (CH₃)₂CH);

24.0 (d, J_C,P= 3.9, (CH₃)₂CH); 24.0 (d, J_C,P= 4.0, (CH₃)₂CH); 47.7 (d, J_C,F= 9.1, CH₂-1ʹ);

65.2 (d, J_C,P= 168.5, CH₂-4ʹ); 71.3 (d, J_C,P= 6.6, CH(CH₃)₂); 71.4 (d, J_C,P= 6.7, CH(CH₃)₂);

77.7 (dd, J_C,F= 18.5, J_C,P= 10.4, CH-2ʹ); 82.0 (d, J_C,F= 173.6, CH₂-3ʹ); 154.4 (C-2); 159.5

(CH-6); 166.5 (C-4). ³¹P{¹H} NMR (202.3 MHz, CDCl₃): 18.80. ¹⁹F NMR (470.4 MHz,

CDCl₃): -229.09 (td, J = 47.1, 22.8).

5.7.2. Diisopropyl (S)- ({1-[(4-amino-1,3,5-triazin-2-yl)oxy]-3-fluoropropan-2-

yl}oxymethyl)phosphonate (8b)

Yield: 201 mg (16 %) of yellowish oil. R_f= 0.63 (S1). IR: (CCl₄) ν_max3552, 3501, 3434,

3328, 3328, 3201, 2981, 2937, 1658, 1636, 1611, 1587, 1577, 1542, 1464, 1455, 1428, 1386,

1376, 1337, 1312, 1245, 1179, 1142, 1106, 1011, 993, 890, 823 cm^-1. ESIMS m/z 367.2 (M +

1)⁺(73). HRMS m/z (ESI) calcd. for C₁₃H₂₅O₅N₄FP (M + 1)⁺367.1541, found 367.1542. ¹H

NMR (500.0 MHz, CDCl₃): 1.310, 1.313, 1.322, 1.323 (4 × d, 4 × 3H, J_vic= 6.1, (CH₃)₂CH);

23

3.94 (dd, 1H, J_gem= 13.6, J_H,P= 8.6, H-4ʹb); 3.98 (dd, 1H, J_gem= 13.6, J_H,P= 8.8, H-4ʹa); 4.06

(m, 1H, H-2ʹ); 4.42 – 4.50 (m, 2H, H-1ʹ); 4.58 (ddd, 1H, J_H,F= 47.1, J_gem= 10.2, J_3ʹb,2ʹ= 5.2,

H-3ʹb); 4.69 (ddd, 1H, J_H,F= 47.1, J_gem= 10.2, J_3ʹa,2ʹ= 3.8, H-3ʹa); 4.75 (m, 2H, CH(CH₃)₂);

5.79 (bs, 2H, NH₂); 8.35 (s, 1H, H-6). ¹³C NMR (125.7 MHz, CDCl₃): 23.9, 23.9 (2 × d, J_C,P

= 4.7, (CH₃)₂CH); 24.1, 24.1 (2 × d, J_C,P= 3.7, (CH₃)₂CH); 64.9 (d, J_C,F= 7.8, CH₂-1ʹ); 65.31

(d, J_C,P= 168.3, CH₂-4ʹ); 71.3, 71.36 (2 × d, J_C,P= 6.6, CH(CH₃)₂); 78.0 (dd, J_C,F= 19.4, J_C,P

= 10.4, CH-2ʹ); 82.3 (d, J_C,F= 172.5, CH₂-3ʹ); 167.9 (C-4); 168.10 (CH-6); 170.1 (C-2);

³¹P{¹H} NMR (202.3 MHz, CDCl₃): 18.97. ¹⁹F NMR (470.4 MHz, CDCl₃): -227.58 (td, J =

47.1, 19.1).

5.8. (S)-{[1-(4-Amino-2-oxo-1,3,5-triazin-1(2H)-yl)-3-fluoropropan-2-

yl]oxymethyl}phosphonic acid (9).

Bromotrimethylsilane (0.9 ml, 6.8 mmol) was added to a solution of 8a (228 mg, 0.6 mmol)

in acetonitrile (15 ml). The mixture was stirred at room temperature overnight, evaporated and

the residue coevaporated with water (2 × 2 ml). The residue was dissolved in water (3 ml) and

applied onto a column of Dowex 1 (acetate form). The ion exchanger was washed with water,

than with a gradient 0-1M acetic acid and finally with 1M formic acid. Appropriate UV

absorbing fractions were evaporated, the residue co-evaporated with methanol (2 x 5 mL) and

toluene (2 x 5 mL), and the product was finally lyophilised from water. Yield: 158 mg (74 %)

of 9 as a white foam. [α]_D-53.3 (c 0.377, H₂O). IR: (KBr) ν_max3385, 3107, 1681, 1616, 1524,

1308, 1203, 1176, 1003, 991 cm^-1. ESIMS m/z 283.0 (M + 1)⁺(47). HRMS m/z (ESI) calcd.

for C₇H₁₃O₅N₄FP (M + 1)⁺283.0602 found 283.0603. Anal. Calcd. for C₇H₁₂FO₅N₄P. 3.3

H₂O: C, 24.6; H, 5.5; N, 16.4; P, 9.1; F, 5.6. Found: C, 24.9; H, 5.2; N, 16.2; P, 8.9; F, 5.3. ¹H

NMR (500.0 MHz, D₂O): 3.69 (dd, 1H, J_gem= 13.7, J_H,P= 9.0, H-4ʹb); 3.85 – 4.07 (m, 4H, H-

1ʹ,2ʹ,4ʹa); 4.59 (ddd, 1H, J_H,F= 46.3, J_gem= 10.8, J_3ʹb,2ʹ= 3.1, H-3ʹb); 4.77 (ddd, 1H, J_H,F

=

24

47.1, J_gem= 10.8, J_3ʹa,2ʹ= 2.8, H-3ʹa); 8.47 (s, 1H, H-6). ¹³C NMR (125.7 MHz, D₂O): 50.2 (d,

_C,F= 8.2, CH₂-1ʹ); 68.0 (d, J_C,P= 158.4, CH₂-4ʹ); 79.4 (dd, J_C,F= 18.6, J_C,P= 9.8, CH-2ʹ);

J

84.5 (d, J_C,F= 168.6, CH₂-3ʹ); 150.6 (C-2); 162.6 (C-4); 165.0 (CH-6). ³¹P{¹H} NMR (202.3

MHz, D₂O): 16.56. ¹⁹F NMR (470.4 MHz, CDCl₃): -228.54 (ddd, J = 47.1, 46.3, 22.9).

5.9. Bis(isopropyl L-phenylalanine) amidate prodrug of (S)-1-[3-fluoro-2-

(phosphonomethoxy)propyl]-5-azacytosine (11).

A mixture of 8a (150 mg, 0.4 mmol), acetonitrile (10 ml) and bromotrimethylsilane (544 ꢀL,

4.1 mmol) was stirred at room temperature overnight, evaporated (under argon), and the

residue was co-evaporated with toluene (2 × 2 ml). The residue and ethyl L-alaninate (252

mg, 1.6 mmol) were dissolved in pyridine (3.3 mL) and triethylamine (0.8 mL). The solution

was warmed to 50 °C and another solution consisting of Aldrithiol-2 (541 mg, 2.5 mmol),

triphenylphosphine and pyridine (4.1 mL) was added. The resulting mixture was heated at 50

°C for 4 h and further stirred at room temperature overnight. The mixture was evaporated, and

the residue was chromatographed on silica gel in a gradient of 0- 15% methanol in

chloroform. Yield: 128 mg (64 %) of 11 as white amorphous solid. [α]_D-55.8 (c 0.251,

CHCl₃). IR: (CHCl₃) ν_max3538, 3487, 3421, 3394, 3199, 2988, 2940, 2877, 1733, 1690, 1463,

1635, 1599, 1507, 1478, 1391, 1377, 1368, 1108, 1062, 1020, 991cm^-1. ESIMS m/z 503.0 (M

+ Na)⁺(100). HRMS m/z (ESI) calcd. for C₁₇H₃₀O₇N₆FNaP (M + Na)⁺503.1790, found

503.1790. Anal. Calcd for C₁₇H₃₀FO₇N₆FP. 1.5 H₂O: C, 40.2; H, 6.6; N, 16.6; P, 6.0; F, 3.7.

Found: C, 40.5; H, 6.3; N, 16.3; P, 6.0; F, 3.7. ¹H NMR (500.0 MHz, CD₃OD): 1.26, 1.29 (2

× t, 2 × 3H, J_vic= 7.1, CH₃CH₂O); 1.39 (dd, 3H, J_vic= 7.2, J_H,P= 1.0, H-3-Ala); 1.40 (d, 3H,

J_vic= 7.2, H-3-Ala); 3.69 (dd, 1H, J_gem= 13.0, J_H,P= 11.5, H-4ʹb); 3.79 (dd, 1H, J_gem= 14.2,

J_1ʹb,2ʹ= 8.2, H-1ʹb); 3.89 – 3.99 (m, 3H, H-2ʹ, H-2-Ala); 4.00 (dd, 1H, J_gem= 13.0, J_H,P= 7.0,

H-4ʹa); 4.09 – 4.25 (m, 5H, H-1ʹa, CH₃CH₂O); 4.51 (ddd, 1H, J_H,F= 47.1, J_gem= 10.7, J_3ʹb,2ʹ

=

25

13

4.4, H-3ʹb); 4.70 (ddd, 1H, J_H,F= 47.6, J_gem= 10.7, J_3ʹa,2ʹ= 3.0, H-3ʹa); 8.26 (s, 1H, H-6). C

NMR (125.7 MHz, CD₃OD): 14.5, 14.5 (CH₃CH₂O); 20.8 (d, J_C,P= 6.9, CH₃-3-Ala); 21.1 (d,

J_C,P= 5.3, CH₃-3-Ala); 48.8 (CH₂-1ʹ); 49.5, 50.3 (CH-2-Ala); 62.4, 62.4 (CH₃CH₂O); 67.3 (d,

J_C,P= 136.6, CH₂-4ʹ); 79.5 (dd, J_C,F= 18.5, J_C,P= 12.8, CH-2ʹ); 83.3 (d, J_C,F= 171.1, CH₂-3ʹ);

157.28 (C-2); 161.4 (CH-6); 168.3 (C-4); 175.7 (d, J_C,P= 2.2, C-1-Ala); 175.8 (d, J_C,P= 4.5,

31

19

C-1-Ala). P{¹H} NMR (202.3 MHz, CD₃OD): 24.66. F NMR (470.4 MHz, CD₃OD): -

229.85 (ddd, J = 47.6, 47.1, 22.4).

5.10. (S)-6-Amino-3-{2-[(diisopropoxyphosphoryl)methoxy]-3-fluoropropyl}-4-oxo-

1,2,3,4-tetrahydro-1,3,5-triazin-1-ium acetate (12).

Compound 8a (351 mg, 0.9 mmol) was hydrogenated in acetic acid (10 mL) on 10%

palladium on charcoal (109 mg) at atmospheric pressure (TLC control in S1) for 24 h. The

mixture was filtered through a Celite pad and the filtrate was evaporated. The residue was co-

evaporated with toluene (2 x 5 mL), methanol (5 mL) and hexane (5 mL). Yield: 351 mg (90

%) of 12 as colourless oil. [α]_D-7.7 (c 0.350, CHCl₃). IR: (CHCl₃) ν_max3320, 3189, 2985,

2748, 2502, 1716, 1654, 1609, 1557, 1467, 1409, 1235, 1179, 1142, 1002, 891, 653, 601 cm^-1.

ESIMS m/z 367.2 (M - 1)^-(100). HRMS m/z (ESI) calcd. for C₁₃H₂₅O₅N₄FP (M - 1)^-

367.1552, found 367.1547. ¹H NMR (500.0 MHz, CDCl₃): 1.31, 1.33, 1.34 (3 × d, 12H, J_vic

=

6.0, (CH₃)₂CH); 1.99 (bs, 3H, CH₃COO); 3.31 (dd, 1H, J_gem= 14.7, J_1ʹb,2ʹ= 7.4, H-1ʹb); 3.63

(dd, 1H, J_gem= 14.7, J_1ʹa,2ʹ= 3.8, H-1ʹa); 3.81 (dd, 1H, J_gem= 13.6, J_H,P= 9.6, H-4ʹb); 3.93 (m,

1H, H-2ʹ); 3.94 (dd, 1H, J_gem= 13.6, J_H,P= 9.0, H-4ʹa); 4.43 (ddd, 1H, J_H,F= 47.2, J_gem= 10.6,

J_3ʹb,2ʹ= 5.0, H-3ʹb); 4.57 (ddd, 1H, J_H,F= 47.2, J_gem= 10.6, J_3ʹa,2ʹ= 2.9, H-3ʹa); 4.61 (d, 1H,

J_gem= 9.6, H-6b); 4.67 – 4.95 (m, 2H, CH(CH₃)₂); 4.79 (d, 1H, J_gem= 9.6, H-6a); 8.88 (bs,

+

4H, NH₂+ NH₂). ¹³C NMR (125.7 MHz, CDCl₃): 22.9 (CH₃COO); 23.9, 24.0 (2 × d, J_C,P

=

4.6, (CH₃)₂CH); 24.0, 24.0 (2 × d, J_C,P= 3.9, (CH₃)₂CH); 46.2 (d, J_C,F= 8.7, CH₂-1ʹ); 57.4

26

(CH₂-6); 64.9 (d, J_C,P= 169.9, CH₂-4ʹ); 71.7, 72.0 (2 × d, J_C,P= 6.8, CH(CH₃)₂); 79.3 (dd, J_C,F

= 18.3, J_C,P= 11.8, CH-2ʹ); 82.5 (d, J_C,F= 172.9, CH₂-3ʹ); 153.6 (C-2); 156.5 (C-4); 177.8

(CH₃COO). ³¹P{¹H} NMR (202.3 MHz, CDCl₃): 19.07. ¹⁹F NMR (470.4 MHz, CDCl₃): -

227.48 (td, J = 47.2, 20.9).

5.11. (S)-1-[3-Fluoro-2-(phosphonomethoxy)propyl]-5,6-dihydro-5-azacytosine (13).

A mixture of 12 (332 mg, 0.8 mmol), acetonitrile (19 ml) and bromotrimethylsilane (1.0 ml,

7.8 mmol) was stirred at room temperature overnight, than evaporated and the residue was co-

evaporated with acetonitrile (2 x 3 mL) and water (2 × 3 ml). The residue was extracted with

ether/water (5 mL/30 mL), ethyl acetate/water (5mL/30 mL) and ether/water (5 mL/30 mL).

The aqueous layer was evaporated, and the residue was co-evaporated with methanol (5 mL),

toluene (2 x 5 mL) and ethyl acetate (5 mL). The final purification was performed on a

reverse phase C18 column (S8) in water, and the product was lyophilised from water. Yield:

180 mg (75 %) of 11 as a yellowish foam. [α]_D-13.8 (c 0.276, D₂O). IR: (KBr) ν_max3409,

2810, 2732, 2700, 2330, 1700, 1623, 1172, 1110, 1063, 1005 cm^-1. ESIMS m/z 285.1 (M +

1)⁺(92). HRMS m/z (ESI) calcd. for C₇H₁₅O₅N₄FP (M + 1)⁺285.0759, found 285.0760.

Anal. Calcd for C₇H₁₄FO₅N₄FP. 9 H₂O: C, 18.8; H, 7.1; N, 12.6; P, 6.9; F, 4.3. Found: C,

18.7; H, 7.0; N, 12.9; P, 6.9; F, 4.0. ¹H NMR (500.0 MHz, D₂O): 3.53 (dd, 1H, J_gem= 15.1,

J_1ʹb,2ʹ= 7.8, H-1ʹb); 3.72 (dd, 1H, J_gem= 15.1, J_1ʹa,2ʹ= 3.8, H-1ʹa); 3.77 (dd, 1H, J_gem= 13.3,

J_H,P= 9.8, H-4ʹb); 3.92 (dd, 1H, J_gem= 13.3, J_H,P= 9.0, H-4ʹa); 3.96 (m, 1H, H-2ʹ); 4.53 (ddd,

1H, J_H,F= 46.6, J_gem= 10.8, J_3ʹb,2ʹ= 4.0, H-3ʹb); 4.69 (ddd, 1H, J_H,F= 47.3, J_gem= 10.8, J_3ʹa,2ʹ

=

3.2, H-3ʹa); 4.80 (d, 1H, J_gem= 10.5, H-6b); 4.89 (d, 1H, J_gem= 10.5, H-6a). ¹³C NMR (125.7

MHz, D₂O): 48.5 (d, J_C,F= 8.0, CH₂-1ʹ); 59.0 (CH₂-6); 68.2 (d, J_C,P= 159.4, CH₂-4ʹ); 81.3

(dd, J_C,F= 18.3, J_C,P= 11.5, CH-2ʹ); 85.1 (d, J_C,F= 167.7, CH₂-3ʹ); 154.3 (C-2); 156.6 (C-4).

27

³¹P{¹H} NMR (202.3 MHz, D₂O): 17.74. ¹⁹F NMR (470.4 MHz, D₂O): -229.37 (ddd, J =

47.3, 46.6, 23.8).

5.12. The reaction of compound 12 with 1-aryl-2-bromoacetophenones (14a-c).

A mixture of 12 (0.51 mmol), bromoacetophenone 14a-c (0.5 mmol) and sodium bicarbonate

(43 mg, 0.5 mmol) in DMF (10 mL) was stirred at room temperature for 24 h. The mixture

was evaporated and the residue was chromatographed on a preparative TLC plate (S7) in the

system mentioned below.

5.12.1. Diisopropyl (S)-{[(1-fluoro-3-(2-oxo-7-phenyl-1,2,8,8a-tetrahydroimidazo[1,2-

a][1,3,5]triazin-3(4H)-yl)propan-2-yl)oxy]methyl}phosphonate (15a).

Yield: 95 mg (40 %) of 15a as yellowish oil. Chromatography in the chloroform-methanol

(15:1) system, R_f= 0.42 (S3). IR: (CCl₄) ν_max3447, 3198, 3156, 3059, 3035, 2981, 2935,

2876, 1692, 1611, 1602, 1561, 1531, 1468, 1468, 1452, 1452, 1386, 1376, 1316, 1282, 1250,

1179, 1142, 1105, 1009, 993, 890, 710 cm^-1. ESIMS m/z 469.3 (M + 1)⁺(92). HRMS m/z

(ESI) calcd. for C₂₁H₃₁O₅N₄FP (M + 1)⁺469.1938, found 469.1943. ¹H NMR (500.0 MHz,

CDCl₃): 1.28, 1.30, 1.31 (3 × d, 12H, J_vic= 6.2, (CH₃)₂CH); 3.42 (dd, 1H, J_gem= 14.7, J_1ʹb,2ʹ

=

7.5, H-1ʹb); 3.78 (dd, 1H, J_gem= 13.5, J_H,P= 9.7, H-4ʹb); 3.87 (dd, 1H, J_gem= 14.7, J_1ʹa,2ʹ= 3.4,

H-1ʹa); 4.01 (dd, 1H, J_gem= 13.5, J_H,P= 8.7, H-4ʹa); 4.05 (m, 1H, H-2ʹ); 4.48 (ddd, 1H, J_H,F

47.4, J_gem= 10.6, J_3ʹb,2ʹ= 5.2, H-3ʹb); 4.66 (ddd, 1H, J_H,F= 47.4, J_gem= 10.6, J_3ʹa,2ʹ= 2.8, H-

=

3ʹa); 4.67 – 4.77 (m, 2H, (CH₃)₂CH); 5.32, 5.61 (2 × d, 2 × 1H, J_gem= 8.9, H-9); 6.94 (s, 1H,

H-7); 7.23 (m, 1H, H-p-Ph); 7.33 – 7.37 (m, 2H, H-m-Ph); 7.67 – 7.71 (m, 2H, H-o-Ph). ¹³C

NMR (125.7 MHz, CDCl₃): 23.9, 24.0 (2 × d, J_C,P= 4.6, (CH₃)₂CH); 24.0, 24.1 (2 × d, J_C,P

3.9, (CH₃)₂CH); 46.8 (d, J_C,F= 9.3, CH₂-1ʹ); 60.4 (CH₂-9); 65.3 (d, J_C,P= 169.4, CH₂-4ʹ);

71.2, 71.5 (2 × d, J_C,P= 6.7, CH(CH₃)₂); 80.1 (dd, J_C,F= 18.1, J_C,P= 11.5, CH-2ʹ); 82.8 (d, J_C,F

= 172.8, CH₂-3ʹ); 107.6 (CH-7); 124.7 (CH-o-Ph); 127.0 (CH-m-Ph); 128.6 (CH-o-Ph); 133.3

=

28

Article Doi

Synthesis of fluorinated acyclic nucleoside phosphonates with 5-azacytosine base moiety

DOI: 10.1016/j.tet.2019.130529

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.tet.2019.130529

Full text of DOI:10.1016/j.tet.2019.130529

Products guided by the article

R&D Labs maybe for 941293-48-7

Relevant to this article

Hot Product