Synthesis of (2′S)‐2′‐Deoxy‐2′‐C‐methylpurine Nucleosides and Their Phosphoramidites

We describe the stereoselective synthesis of (2′S)‐2′‐deoxy‐2′‐C‐methyladenosine ( 12 ) and (2′S)‐2′‐deoxy‐2′‐C‐methylinosine ( 14 ) as well as their corresponding cyanoethyl phosphoramidites 16 and 19 from 6‐O‐(2,6‐dichlorophenyl)inosine as starting material. The methyl group at the 2′‐position was introduced via a Wittig reaction (→ 3 , Scheme 1) followed by a stereoselective oxidation with OsO₄ (→ 4 , Scheme 2). The primary‐alcohol moiety of 4 was tosylated (→ 5 ) and regioselectively reduced with NaBH₄ (→ 6 ). Subsequent reduction of the 2′‐alcohol moiety with Bu₃SnH yielded stereoselectively the corresponding (2′S)‐2′‐deoxy‐2′‐C‐methylnucleoside (→ 8a ).     

7‐Halogenated 7‐Deazapurine 2′‐Deoxyribonucleosides Related to 2′‐Deoxyadenosine, 2′‐Deoxyxanthosine,and 2′‐Deoxyisoguanosine: Syntheses and Properties

A series of 7‐fluorinated 7‐deazapurine 2′‐deoxyribonucleosides related to 2′‐deoxyadenosine, 2′‐deoxyxanthosine, and 2′‐deoxyisoguanosine as well as intermediates 4b – 7b, 8, 9b, 10b , and 17b were synthesized. The 7‐fluoro substituent was introduced in 2,6‐dichloro‐7‐deaza‐9H‐purine ( 11a ) with Selectfluor (Scheme 1). Apart from 2,6‐dichloro‐7‐fluoro‐7‐deaza‐9H‐purine ( 11b ), the 7‐chloro compound 11c was formed as by‐product. The mixture 11b / 11c was used for the glycosylation reaction; the separation of the 7‐fluoro from the 7‐chloro compound was performed on the level of the unprotected nucleosides. Other halogen substituents were introduced with N‐halogenosuccinimides ( 11a → 11c – 11e ). Nucleobase‐anion glycosylation afforded the nucleoside intermediates 13a – 13e (Scheme 2). The 7‐fluoro‐ and the 7‐chloro‐7‐deaza‐2′‐deoxyxanthosines, 5b and 5c , respectively, were obtained from the corresponding MeO compounds 17b and 17c , or 18 (Scheme 6). The 2′‐deoxyisoguanosine derivative 4b was prepared from 2‐chloro‐7‐fluoro‐7‐deaza‐2′‐deoxyadenosine 6b via a photochemically induced nucleophilic displacement reaction (Scheme 5). The pK_a values of the halogenated nucleosides were determined (Table 3). ¹³C‐NMR Chemical‐shift dependencies of C(7), C(5), and C(8) were related to the electronegativity of the 7‐halogen substituents (Fig. 3). In aqueous solution, 7‐halogenated 2′‐deoxyribonucleosides show an approximately 70% S population (Fig. 2 and Table 1).     

Stereoselective Synthesis of [5‐[4,4,4,4′,4′,4′‐Hexafluoro‐N‐(2‐hydroxyethoxy)‐D‐valine]]‐ and [5‐[4,4,4,4′,4′,4′‐Hexafluoro‐N‐(2‐hydroxyethoxy)‐L‐valine]cyclosporin A

Addition of various amines to the 3,3‐bis(trifluoromethyl)acrylamides 10a and 10b gave the tripeptides 11a – 11f , mostly as mixtures of epimers (Scheme 3). The crystalline tripeptide 11f ₂ was found to be the N‐terminal (2‐hydroxyethoxy)‐substituted (R,S,S)‐ester HOCH₂CH₂O‐D ‐Val(F₆)‐MeLeu‐Ala‐O^tBu by X‐ray crystallography. The C‐terminal‐protected tripeptide 11f ₂ was condensed with the N‐terminus octapeptide 2b to the depsipeptide 12a which was thermally rearranged to the undecapeptide 13a (Scheme 4). The condensation of the epimeric tripeptide 11f ₁ with the octapeptide 2b gave the undecapeptide 13b directly. The undecapeptides 13a and 13b were fully deprotected and cyclized to the [5‐[4,4,4,4′,4′,4′‐hexafluoro‐N‐(2‐hydroxyethoxy)‐D ‐valine]]‐ and [5‐[4,4,4,4′,4′,4′‐hexafluoro‐N‐(2‐hydroxyethoxy)‐L ‐valine]]cyclosporins 14a and 14b , respectively (Scheme 5). Rate differences observed for the thermal rearrangements of 12a to 13a and of 12b to 13b are discussed.     

Novel Fluoride‐Labile Nucleobase‐Protecting Groups for the Synthesis of 3′(2′)‐O‐Aminoacylated RNA Sequences

With the aim to develop a general approach to a total synthesis of aminoacylated t‐RNAs and analogues, we describe the synthesis of stabilized, aminoacylated RNA fragments, which, upon ligation, could lead to aminoacylated t‐RNA structures. Novel RNA phosphoramidites with fluoride‐labile 2′‐O‐[(triisopropylsilyl)oxy]methyl (=tom) sugar‐protecting and N‐{{2‐[(triisopropylsilyl)oxy]benzyl}oxy}carbonyl (=tboc) base‐protecting groups were prepared (Schemes 4 and 5), as well as a solid support containing an immobilized N⁶‐tboc‐protected adenosine with an orthogonal (photolabile) 2′‐O‐[(S)‐1‐(2‐nitrophenyl)ethoxy]methyl (=(S)‐npeom) group (Scheme 6). From these building blocks, a hexameric oligoribonucleotide was prepared by automated synthesis under standard conditions (Scheme 7). After the detachment from the solid support, the resulting fully protected sequence 34 was aminoacylated with L ‐phenylalanine derivatives carrying photolabile N‐protecting groups (→ 42 and 43 ; Scheme 9). Upon removal of the fluoride‐labile sugar‐ and nucleobase‐protecting groups, the still stabilized, partially with the photolabile group protected precursors 44 and 45 , respectively, of an aminoacylated RNA sequence were obtained (Scheme 9 and Fig. 3). Photolysis of 45 under mild conditions resulted in the efficient formation of the 3′(2′)‐O‐aminoacylated RNA sequence 46 (Fig. 4). Additionally, we carried out model investigations concerning the stability of ester bonds of aminoacylated ribonucleotide derivatives under acidic conditions (Table) and established conditions for the purification and handling of 3′(2′)‐O‐aminoacylated RNA sequences and their stabilized precursors.     

Synthesis,Crystal Structure and Hydrolysis of a Novel Anhydrogalactosucrose: 2′‐Methoxyl‐O‐1′,4′:3′,6′‐dianhydro‐β‐D‐fructofuranosyl 3,6‐anhydro‐4‐chloro‐4‐deoxy‐α‐D‐galactopyranoside

A novel anhydrogalactosucrose derivative 2′‐methoxyl‐O‐1′,4′:3′,6′‐dianhydro‐β‐D‐fructofuranosyl 3,6‐anhydro‐4‐chloro‐4‐deoxy‐α‐D‐galactopyranoside ( 4 ) was prepared from 3,6:1′,4′:3′,6′‐trianhydro‐4‐chloro‐4‐deoxy‐galactosucrose ( 3 ) via a facile method and characterized by ¹H NMR, ¹³C NMR and 2D NMR spectra. The single crystal X‐ray diffraction analysis shows that the title molecule forms a two thee‐dimensional network structure by two kinds of hydrogen bond interactions [O(2) H(2)···O(7), O(5) H(5)···O(8)]. Its stability was investigated by acid hydrolysis reaction treated with sulfuric acid, together with the formation of 1,6‐Di‐O‐methoxy‐4‐chloro‐4‐deoxy‐β‐D‐galactopyranose ( 5 ) and 2,2‐Di‐C‐methoxy‐1,4:3,6‐dianhydromannitol ( 6 ). According to the result, the relative stability of the ether bonds in the structure is in the order: C(1) O C(5)≈C(3′) O C(6′)≈C(1′) O C(4′)>C(3) O C(6)≈C(1) O C(2′)>C(2′) O C(5′).     

Asymmetric Hydroformylation of Vinylfurans Catalyzed by {(11bS)‐4‐{[(1R)‐2′‐Phosphino[1,1′‐binaphthalen]‐2‐yl]oxy}dinaphtho[2,1‐d:1′,2′‐f]‐ [1,3,2]dioxaphosphepin}rhodium(I) [RhI{(R,S)‐binaphos}] Derivatives

The asymmetric hydroformylation of 2‐ and 3‐vinylfurans ( 2a and 2b , resp.) was investigated by using [Rh{(R,S)‐binaphos}] complexes as catalysts ((R,S)‐binaphos = (11bS)‐4‐{[1R)‐2′‐phosphino[1,1′‐binaphthalen]‐2‐yl]oxy}dinaphtho[2,1‐d:1′,2′‐f][1,3,2]dioxaphosphepin; 1 ). Hydroformylation of 2 gave isoaldehydes 3 in high regio‐ and enantioselectivities (Scheme 2 and Table). Reduction of the aldehydes 3 with NaBH₄ successfully afforded the corresponding alcohols 5 without loss of enantiomeric purity (Scheme 3).     

Flip‐type disorder in 3‐substituted 2,2′:5′,2′′‐terthiophenes

In the crystal structures of four thiophene derivatives, (E)‐3′‐[2‐(anthracen‐9‐yl)ethenyl]‐2,2′:5′,2′′‐terthiophene, C₂₈H₁₈S₃, (E)‐3′‐[2‐(1‐pyrenyl)ethenyl]‐2,2′:5′,2′′‐terthiophene, C₃₀H₁₈S₃, (E)‐3′‐[2‐(3,4‐dimethoxyphenyl)ethenyl]‐2,2′:5′,2′′‐terthiophene, C₂₂H₁₈O₂S₃, and (E,E)‐1,4‐bis[2‐(2,2′:5′,2′′‐terthiophen‐3′‐yl)ethenyl]‐2,5‐dimethoxybenzene, C₃₆H₂₆O₂S₆, at least one of the terminal thiophene rings is disordered and the disorder is of the flip type. The terthiophene fragments are far from being coplanar, contrary to terthiophene itself. The central C—C=C—C fragments are almost planar but the bond lengths suggest slight delocalization within this fragment. The crystal packing is determined by van der Waals interactions and some weak, relatively short, C—H...S and C—H...π directional contacts.     

Synthesis and Pairing Properties of 3′‐Deoxyribopyranose (4′→2′)‐Oligonucleotides (‘p‐DNA')

The preparation and the pairing properties of the new 3′‐deoxyribopyranose (4′→2′)‐oligonucleotide (=p‐DNA) pairing system, based on 3′‐deoxy‐β‐D ‐ribopyranose nucleosides is presented. D ‐Xylose was efficiently converted to the prefunctionalized 3‐deoxyribopyranose derivative 4‐O‐[(tert‐butyl)dimethylsilyl]‐3‐deoxy‐D ‐ribopyranose 1,2‐diacetate 8 (obtained as a 4 : 1 mixture of α‐ and β‐D ‐anomers; Scheme 1). From this sugar building block, the corresponding, appropriately protected thymine, guanine, 5‐methylcytosine, and purine‐2,6‐diamine nucleoside phosphoramidites 29 – 32 were prepared in a minimal number of steps (Schemes 2–4). These building blocks were assembled on a DNA synthesizer, and the corresponding p‐DNA oligonucleotides were obtained in good yields after a one‐step deprotection under standard conditions, followed by HPLC purification (Scheme 5 and Table 1). Qualitatively, p‐DNA shows the same pairing behavior as p‐RNA, forming antiparallel, exclusively Watson‐Crick‐paired duplexes that are much stronger than corresponding DNA duplexes. Duplex stabilities within the three related (i.e., based on ribopyranose nucleosides) oligonucleotide systems p‐RNA, p‐DNA, and 3′‐O‐Me‐p‐RNA were compared with each other (Table 2). Intrinsically, p‐RNA forms the strongest duplexes, followed by p‐DNA, and 3′‐O‐Me‐p‐RNA. However, by introducing the nucleobases purine‐2,6‐diamine (D) and 5‐methylcytosine (M) instead of adenine and cytosine, a substantial increase in stability of corresponding p‐DNA duplexes was observed.     

Oligonucleotides Containing Novel 4′‐C‐ or 3′‐C‐(Aminoalkyl)‐Branched Thymidines

The synthesis of four novel 3′‐C‐branched and 4′‐C‐branched nucleosides and their transformation into the corresponding 3′‐O‐phosphoramidite building blocks for automated oligonucleotide synthesis is reported. The 4′‐C‐branched key intermediate 11 was synthesized by a convergent strategy and converted to its 2′‐O‐methyl and 2′‐deoxy‐2′‐fluoro derivatives, leading to the preparation of novel oligonucleotide analogues containing 4′‐C‐(aminomethyl)‐2′‐O‐methyl monomer X and 4′‐C‐(aminomethyl)‐2′‐deoxy‐2′‐fluoro monomer Y (Schemes 2 and 3). In general, increased binding affinity towards complementary single‐stranded DNA and RNA was obtained with these analogues compared to the unmodified references (Table 1). The presence of monomer X or monomer Y in a 2′‐O‐methyl‐RNA oligonucleotide had a negative effect on the binding affinity of the 2′‐O‐methyl‐RNA oligonucleotide towards DNA and RNA. Starting from the 3′‐C‐allyl derivative 28 , 3′‐C‐(3‐aminopropyl)‐protected nucleosides and 3′‐O‐phosphoramidite derivatives were synthesized, leading to novel oligonucleotide analogues containing 3′‐C‐(3‐aminopropyl)thymidine monomer Z or the corresponding 3′‐C‐(3‐aminopropyl)‐2′‐O,5‐dimethyluridine monomer W (Schemes 4 and 5). Incorporation of the 2′‐deoxy monomer Z induced no significant changes in the binding affinity towards DNA but decreased binding affinity towards RNA, while the 2′‐O‐methyl monomer Z induced decreased binding affinity towards DNA as well as RNA complements (Table 2).     

Synthesis of 3′‐Thioamido‐Modified 3′‐Deoxythymidine 5′‐Triphosphates by Regioselective Thionation and Their Use as Chain Terminators in DNA Sequencing

The thioamide derivatives 3′‐deoxy‐5′‐O‐(4,4′‐dimethoxytrityl)‐3′‐[(2‐methyl‐1‐thioxopropyl)amino]thymidine ( 4a ) and 3′‐deoxy‐5′‐O‐(4,4′‐dimethoxytrityl)‐3′‐{{6‐{[(9H‐(fluoren‐9‐ylmethoxy)carbonyl]amino}‐1‐thioxohexyl}amino}thymidine ( 4b ) were synthesized by regioselective thionation of the corresponding amides 3a and 3b with 2,4‐bis(4‐methoxyphenyl)‐1,3,2,4‐dithiadiphosphetane 2,4‐disulfide (Lawesson's reagent). The addition of exact amounts of pyridine to the reaction mixture proved to be essential for an efficient transformation. The thioamides were converted into the corresponding 5′‐triphosphates 6a and 6b . Compound 6a was chosen for DNA sequencing experiments, and 6b was further labelled with fluorescein (→ 8 ).     

Functionalized 3,3′,5,5′‐Tetraaryl‐1,1′‐Biphenyls: Novel Platforms for Molecular Receptors

This paper describes the development of novel aromatic platforms for supramolecular construction. By the Suzuki cross‐coupling protocol, a variety of functionalized m‐terphenyl derivatives were prepared (Schemes 1–4). Macrolactamization of bis(ammonium salt) (S,S)‐ 6 with bis(acyl halide) 7 afforded the macrocyclic receptor (S,S)‐ 2 (Scheme 1), which was shown by ¹H‐NMR titration studies to form ‘nesting' complexes of moderate stability (K_a between 130 and 290 M ^?1, 300 K) with octyl glucosides 13 – 15 (Fig. 2) in the noncompetitive solvent CDCl₃. Suzuki cross‐coupling starting from 3,3′,5,5′‐tetrabromo‐1,1′‐biphenyl provided access to a novel series of extended aromatic platforms (Scheme 5) for cleft‐type (Fig. 1) and macrotricyclic receptors such as (S,S,S,S)‐ 1 . Although mass‐spectral evidence for the formation of (S,S,S,S)‐ 1 by macrolactamization between the two functionalized 3,3′,5,5′‐tetraaryl‐1,1′‐biphenyl derivatives (S,S)‐ 33 and 36 was obtained, the ¹H‐ and ¹³C‐NMR spectra of purified material remained rather inconclusive with respect to both purity and constitution. The versatile access to the novel, differentially functionalized 3,3′,5,5′‐tetrabromo‐1,1′‐biphenyl platforms should ensure their wide use in future supramolecular construction.     

Carotinoide mit 7-Oxabicyclo[2.2.1]heptyl-Endgruppen. Teil I. Versuch einer Synthese von Cycloviolaxanthin (= (3S,5R,6R,3′S,5′R,6′R)-3,6:3′,6′-Diepoxy-5,6,5′,6′-tetrahydro-β,β-carotin-5,5′-diol)

Carotenoids with 7-Oxabicyclo[2,2.1]heptyl End Groups. Attempted Synthesis of Cycloviolaxanthin ( = (3S,5R,6S,3′S,5′R,6′R)-3,6:3′,6′- Diepoxy-5,6,5′,6′-tetrahydro-β,β-carotin-5,5′-diol) Starting from our recently described synthon (+)- 24 , the enantiomerically pure 3,6:4,5:3′,6′:4′,5′-tetraepoxy-4,5,4′,5′-tetrahydro-ε,ε-carotene ( 34 ) and its 15,15′-didehydro analogue 32 were synthesized in eleven and nine steps, respectively (Scheme 4). Chiroptical data show, in contrast to the parent ε,ε-carotene, a very weak interaction between the chiral centers at C(5), C(5′), C(6), C(6′), and the polyene system. Diisobutylaluminium hydride reduction of 32 lead rather than to the expected 15,15′-didehydro analogue 35 of Cycloviolaxanthin ( 8 ), to the polyenyne 36 (Scheme 5). We explain this reaction by an oxirane rearrangement leading to a cyclopropyl ether followed by a fragmentation to an aldehyd on the one side and an enol ether on the other (Scheme 6). This complex rearrangement includes a shift of the whole polyenyne chain from C(6), C(6′) to C(5), C(5′) of the original molecule.     

Aza-Claisen Rearrangement: Synthesis of 5′-branched 5′-aminothymidines

The syntheses of both diastereoisomers of 5′-ethyl-substituted thymidine dimers, the (5′R)- and (5′S)-configurated 33a and 33b respectively, in which the natural phosphodiester linkage is replaced by an amide group (C(3′)-CH₂CONH-CH(5′)(Et)), arc described. Their fully protected derivatives 35a and 35b , respectively, are suitable for incorporation into antisense oligonucleotides. Unexpectedly, an attempted Pd^II-catalysed aza-Claisen rearrangement of trichloroacetimidate 7 provided the diastereoisomerically pure cyclopropane derivative 17 , whose structure was confirmed by X-ray analysis.     

Carbocyclic 5′‐norcytidine (5′‐norcarbodine)

A preparation of (1′R,2′S,3′R,4′S)‐1‐(2′,3′,4′‐trihydroxycyclopent‐1′‐yl)‐lH‐cytosine (5′‐norcarbodine, 3 ) has formally been achieved in 2 steps from (+)‐(1R,4S)‐4‐hydroxy‐2‐cyclopenten‐1‐yl acetate ( 4 ) and cytosine. The L‐like enantiomer of 3 (that is, 6 ) is also reported using the enantiomer of 4 (that is, 7 ). In evalu ating 3 and 6 for antiviral potential against a number of viruses, compound 3 was found to have activity towards Epstein‐Barr virus (EBV).     

2′‐Deoxy‐5‐methylisocytidine

In the title compound, 2‐amino‐1‐(2‐deoxy‐β‐d ‐erythro‐pento­furan­osyl)‐5‐methyl­pyrimidin‐4(1H)‐one, C₁₀H₁₅N₃O₄, the conformation of the N‐glycosidic bond is syn and the 2‐deoxy­ribo­furan­ose moiety adopts an unusual ^OT₁ sugar pucker. The orientation of the exocyclic C4′—C5′ bond is +sc (+gauche).     

Concise Synthesis of [1,1′‐Biisoquinoline]‐4,4′‐diol via a Protecting Group Strategy and Its Application for Potential Liquid‐Crystalline Compounds

The [1,1′‐biisoquinoline]‐4,4′‐diol ( 4a ), which was obtained as hydrochloride 4a ?2 HCl in two steps starting from the methoxymethyl (MOM)‐protected 1‐chloroisoquinoline 8 (Scheme 3), opens access to further O‐functionalized biisoquinoline derivatives. Compound 4a ?2 HCl was esterified with 4‐(hexadecyloxy)benzoyl chloride ( 5b ) to give the corresponding diester 3b (Scheme 4), which could not be obtained by Ni‐mediated homocoupling of 6b (Scheme 2). The ether derivative 2b was accessible in good yield by reaction of 4a ?2 HCl with the respective alkyl bromide 9 under the conditions of Williamson etherification (Scheme 4). Slightly modified conditions were applied to the esterification of 4a ?2 HCl with galloyl chlorides 10a – h as well as etherification of 4a ?2 HCl with 6‐bromohexyl tris(alkyloxy)benzoates 11b , d – h and [(6‐bromohexyl)oxy]‐substituted pentakis(alkyloxy)triphenylenes 14a – c (Scheme 5). Despite the bulky substituents, the respective target 1,1′‐biisoquinolines 12, 13 , and 15 were isolated in 14–86% yield (Table).     

A Novel,One‐Pot Four‐Component Route to 2′‐Thioxo‐2′,3′‐dihydrospiro[indole‐3,6′‐[1,3]thiazin]‐2‐one Derivatives

An efficient route to 2′,3′‐dihydro‐2′‐thioxospiro[indole‐3,6′‐[1,3]thiazin]‐2(1H)‐one derivatives is described. It involves the reaction of isatine, 1‐phenyl‐2‐(1,1,1‐triphenyl‐λ⁵‐phosphanylidene)ethan‐1‐one, and different amines in the presence of CS₂ in dry MeOH at reflux (Scheme 1). The alkyl carbamodithioate, which results from the addition of the amine to CS₂, is added to the α,β‐unsaturated ketone, resulting from the reaction between 1‐phenyl‐2‐(1,1,1‐triphenyl‐λ⁵‐phosphanylidene)ethan‐1‐one and isatine, to produce the 3′‐alkyl‐2′,3′‐dihydro‐4′‐phenyl‐2′‐thioxospiro[indole‐3,6′‐[1,3]thiazin]‐2(1H)‐one derivatives in excellent yields (Scheme 2). Their structures were corroborated spectroscopically (IR, ¹H‐ and ¹³C‐NMR, and EI‐MS) and by elemental analyses.     

2′‐Deoxy‐5‐propynyluridine: a nucleoside with two conformations in the asymmetric unit

The title compound, 1‐(2‐deoxy‐β‐d ‐erythro‐pentofuranosyl)‐5‐(prop‐1‐ynyl)pyrimidin‐2,4(1H,3H)‐dione, C₁₂H₁₄N₂O₅, shows two conformations in the crystalline state: conformer 1 adopts a C2′‐endo (close to ²E; S‐type) sugar pucker and an anti nucleobase orientation [χ = −134.04 (19)°], while conformer 2 shows an S sugar pucker (twisted C2′‐endo–C3′‐exo), which is accompanied by a different anti base orientation [χ = −162.79 (17)°]. Both molecules show a +sc (gauche, gauche) conformation at the exocyclic C4′—C5′ bond and a coplanar orientation of the propynyl group with respect to the pyrimidine ring. The extended structure is a three‐dimensional hydrogen‐bond network involving intermolecular N—H...O and O—H...O hydrogen bonds. Only O atoms function as H‐atom acceptor sites.     

The α‐d anomer of 5‐aza‐7‐deaza‐2′‐deoxyguanosine

In the monohydrate of 2‐amino‐8‐(2‐deoxy‐α‐d ‐erythro‐pento­furan­osyl)‐8H‐imidazo­[1,2‐a]­[1,3,5]­triazin‐4‐one, C₁₀H₁₃N₅O₄·H₂O, denoted (I) or αZ_d, the conformation of the N‐gly­cosyl­ic bond is in the high‐anti range [χ = 87.5 (3)°]. The 2′‐deoxy­ribo­furan­ose moiety adopts a C2′‐endo,C3′‐exo(^2′T_3′) sugar puckering (S‐type sugar) and the conformation at the C4′—C5′ bond is ?sc (trans).     

Synthesis of 9‐Halogenated 9‐Deazaguanine N7‐(2′‐Deoxyribonucleosides)

The syntheses of N⁷‐glycosylated 9‐deazaguanine 1a as well as of its 9‐bromo and 9‐iodo derivatives 1b , c are described. The regioselective 9‐halogenation with N‐bromosuccinimide (NBS) and N‐iodosuccinimide (NIS) was accomplished at the protected nucleobase 4a (2‐{[(dimethylamino)methylidene]amino}‐3,5‐dihydro‐3‐[(pivaloyloxy)methyl]‐4H‐pyrrolo[3,2‐d]pyrimidin‐4‐one). Nucleobase‐anion glycosylation of 4a – c with 2‐deoxy‐3,5‐di‐O‐(p‐toluoyl)‐α‐D ‐erythro‐pentofuranosyl chloride ( 5 ) furnished the fully protected intermediates 6a – c (Scheme 2). They were deprotected with 0.01M NaOMe yielding the sugar‐deprotected derivatives 8a – c (Scheme 3). At higher concentrations (0.1M NaOMe), also the pivaloyloxymethyl group was removed to give 7a – c , while conc. aq. NH₃ solution furnished the nucleosides 1a – c . In D₂O, the sugar conformation was always biased towards S (67–61%).