9H-Purin-6-amine, 9-(5-deoxy-alpha-D-xylofuranosyl)-

• CAS: 72-90-2
• Molecular Formula: C10H13N5O3
• Molecular Weight: 251.24 g/mol
• SMILES: CC1C(C(C(O1)N2C=NC3=C(N=CN=C32)N)O)O
• InChI: InChI=1S/C10H13N5O3/c1-4-6(16)7(17)10(18-4)15-3-14-5-8(11)12-2-13-9(5)15/h2-4,6-7,10,16-17H,1H3,(H2,11,12,13)
• InChIKey: XGYIMTFOTBMPFP-UHFFFAOYSA-N

CAS 72-90-2 delivers a triple-modified α-nucleoside scaffold (α-D-xylofuranosyl, 5′-deoxy) unavailable from adenosine, 5′-deoxyadenosine, or xylo-A. • LigA inhibitor development: validated Ki = 72.8 µM, 9-fold selectivity over human LigI. • ADA enzymology: deconvolutes anomeric, sugar, and 5′-OH contributions. • Phosphorylation-resistant probe for 'signaling-only' adenosine receptor studies.

9-(5-Deoxy-α-D-xylofuranosyl)adenine: Structural Identity and Classification

9H-Purin-6-amine, 9-(5-deoxy-alpha-D-xylofuranosyl)- (CAS 72-90-2) is a purine nucleoside analogue in which adenine is linked via an α-N9-glycosidic bond to a 5-deoxy-α-D-xylofuranose sugar [1]. With a molecular formula of C₁₀H₁₃N₅O₃ and molecular weight of 251.24 g/mol, it differs from adenosine and 5'-deoxyadenosine in three key structural parameters: the anomeric configuration (α vs. natural β), the sugar identity (xylofuranose, a 3'-epimer of ribofuranose), and the absence of a 5'-hydroxyl group . This compound serves as a versatile synthon for constructing modified nucleosides, nucleotides, and glycosylamine-based enzyme inhibitors, and is catalogued in authoritative chemical databases including PubChem and ChEBI [2].

• [1] Mizuno, Y. et al. Nucleic Acid Related Compounds. J. Am. Chem. Soc., 1972, 94, 4737–4739. View Source
• [2] ChEBI. 9-β-D-xylofuranosyladenine (CHEBI:46515) and related entries. EMBL-EBI. View Source

Why Adenosine Analogs Cannot Substitute for This Compound

The compound integrates three orthogonal structural modifications—α-anomeric linkage, xylofuranose sugar (3′-epimer of ribose), and 5′-deoxygenation—into a single scaffold. Each modification independently alters enzyme recognition: the α-configuration changes the spatial presentation of the nucleobase relative to the sugar ring [1]; the xylofuranose 3′-stereochemistry inverts the C3′-OH orientation, affecting hydrogen-bonding networks in enzyme active sites [2]; and the 5′-deoxy modification eliminates the primary alcohol required for 5′-phosphorylation by adenosine kinase and other nucleotide-forming enzymes [3]. The closest comparators—adenosine (β-D-ribofuranose, 5′-OH), 5′-deoxyadenosine (β-D-ribofuranose, 5′-deoxy), xylo-A (β-D-xylofuranose, 5′-OH), and cordycepin (3′-deoxy-β-D-ribofuranose)—each capture only one or two of these features, resulting in distinct metabolic fates, enzyme inhibition profiles, and molecular recognition properties that cannot be conflated. Substituting any of these analogs for CAS 72-90-2 in experiments targeting α-specific glycosylamine recognition, M. tuberculosis LigA inhibition, or conformationally constrained nucleoside enzymology would yield non-comparable or misleading results.

• [1] Gosselin, G.; Bergogne, M.C.; de Rudder, J.; De Clercq, E.; Imbach, J.L. Systematic synthesis and biological evaluation of α- and β-D-xylofuranosyl nucleosides of the five naturally occurring bases in nucleic acids and related analogues. J. Med. Chem., 1986, 29, 203–213. View Source
• [2] Shah, R.H.; Schaeffer, H.J.; Murray, D.H. Enzyme Inhibitors IV: Syntheses of 6-Substituted-9-(5-deoxy-β-D-xylofuranosyl)purines and Their Evaluation as Inhibitors of Adenosine Deaminase. J. Pharm. Sci., 1965, 54, 15–20. View Source
• [3] Long, M.C. et al. Structure-activity relationship for adenosine kinase from Mycobacterium tuberculosis II: Modifications to the ribofuranosyl moiety. Biochem. Pharmacol., 2008, 75, 1568–1581. View Source

Quantitative Differentiation from Closest Structural Analogs

M. tuberculosis LigA Inhibition: Selective Binding Requires 5-Deoxy-α-D-xylofuranosyl Moiety

The α-D-xylofuranosyl configuration with 5′-deoxygenation is essential for selective inhibition of Mycobacterium tuberculosis NAD⁺-dependent DNA ligase (MtuLigA) over human ATP-dependent DNA ligase I. Bis-(5-deoxy-α-D-xylofuranosylated) diamines with a 3-carbon aminoalkyl spacer (Compound 1) achieved a 9-fold discrimination between MtuLigA and human LigI, with a Ki of 72.8 µM for MtuLigA and competitive binding versus NAD⁺ [1]. In contrast, the compound with a 10-carbon spacer (Compound 2) lost this selectivity and bound ATP-dependent enzymes with higher affinity, demonstrating that stereochemical and conformational parameters inherent to the 5-deoxy-α-D-xylofuranosyl head group—not merely linker length—govern target discrimination [1]. Neither adenosine, 5′-deoxyadenosine (β-D-ribo), nor xylo-A (β-D-xylo, 5′-OH) can serve as the recognition element in this pharmacophore because each lacks the precise combination of α-anomeric configuration, xylofuranose 3′-stereochemistry, and 5′-deoxy modification required for LigA active-site complementarity.

Evidence Dimension	Enzyme inhibition potency and selectivity: MtuLigA vs. Human DNA Ligase I
Target Compound Data	Bis-(5-deoxy-α-D-xylofuranosylated) diamine (Compound 1): Ki = 72.8 µM (MtuLigA); 9-fold selectivity over human LigI; competitive with NAD⁺ (Km for NAD⁺ = 1.56 µM)
Comparator Or Baseline	Compound 2 (10-carbon spacer bis-xylofuranosylated diamine): lost selectivity, higher affinity for human LigI; Compound 4 (1,3-phenylene carbamoyl spacer): Ki = 98.3 µM (MtuLigA), also selective but distinct spacer chemotype
Quantified Difference	9-fold selectivity window (Compound 1; MtuLigA vs. human LigI); ~2-fold selectivity for Compound 3; selectivity abolished for Compound 2
Conditions	In vitro full-length enzyme inhibition assay; NAD⁺ concentration 0–50 µM; 10% Me₂SO; 25°C; IC₅₀ determined by relative ligation activity (gel-based); Ki by Lineweaver–Burk analysis

Why This Matters

For anti-tubercular drug development programs, the 5-deoxy-α-D-xylofuranosyl head group provides a validated, selectivity-conferring scaffold; procurement of the free nucleoside (CAS 72-90-2) enables in-house elaboration of focused diamine libraries targeting the NAD⁺ binding pocket of bacterial LigA enzymes.

• [1] Srivastava, S.K. et al. Mycobacterium tuberculosis NAD⁺-dependent DNA ligase is selectively inhibited by glycosylamines compared with human DNA ligase I. Nucleic Acids Res., 2005, 33, 7090–7101. View Source

Adenosine Deaminase Dual Substrate/Inhibitor Behavior

The 6-amino-9-(5-deoxy-β-D-xylofuranosyl)purine (the β-anomer of CAS 72-90-2) is one of the few adenine nucleoside analogs that simultaneously acts as both a substrate and an inhibitor of adenosine deaminase (ADA) [1]. In the Shah et al. (1965) study, the 6-amino analog (corresponding to the target compound's β-anomer) was deaminated by ADA, yet 6-substituted-9-(5-deoxy-β-D-xylofuranosyl)purines as a class also inhibited ADA to varying degrees depending on the 6-substituent [1]. By contrast, adenosine is solely a substrate (Km ≈ 11.9 µM for mammalian ADA; Vmax = 6.8 × 10⁻⁹ M/s) [2], and 5′-deoxyadenosine is deaminated with nearly identical kcat/Km to adenosine by Aspergillus oryzae ADA but does not exhibit inhibitory activity [3]. The α-anomer (CAS 72-90-2) is predicted to display further altered ADA interaction kinetics because the α-configuration changes the glycosidic torsion angle, a parameter known to critically influence ADA binding and catalysis [4].

Evidence Dimension	Adenosine deaminase substrate and inhibitor activity
Target Compound Data	6-Amino-9-(5-deoxy-β-D-xylofuranosyl)purine (β-anomer): acts as ADA substrate (deaminated) and as inhibitor (extent varies with 6-substituent) [Shah 1965]. α-Anomer (CAS 72-90-2): predicted altered kinetics due to α-configuration; no published ADA Ki available.
Comparator Or Baseline	Adenosine: Km = 11.85 µM, Vmax = 6.8 × 10⁻⁹ M/s (mammalian ADA); 5′-Deoxyadenosine: kcat/Km ≈ identical to adenosine (A. oryzae ADA); neither exhibits inhibitory activity
Quantified Difference	Qualitative difference: dual substrate/inhibitor behavior unique to the 5-deoxy-xylofuranosyl scaffold. Quantitative Ki/IC₅₀ values for CAS 72-90-2: not yet published.
Conditions	ADA from calf intestinal mucosa (Shah 1965); mammalian ADA (substrate specificity study); A. oryzae ADA (5′-deoxyadenosine kinetics)

Why This Matters

The dual substrate-inhibitor phenotype of the 5-deoxy-xylofuranosyl scaffold makes CAS 72-90-2 a valuable probe for dissecting ADA transition-state architecture and for developing ADA-resistant adenosine analogs—a property not achievable with adenosine, 5′-deoxyadenosine, or xylo-A alone.

• [1] Shah, R.H.; Schaeffer, H.J.; Murray, D.H. Enzyme Inhibitors IV: Syntheses of 6-Substituted-9-(5-deoxy-β-D-xylofuranosyl)purines and Their Evaluation as Inhibitors of Adenosine Deaminase. J. Pharm. Sci., 1965, 54, 15–20. View Source
• [2] Substrate specificity of adenosine deaminase: the role of the substituents at the 2′- and 3′-carbons of adenine nucleosides, of their configuration and of the conformation of the furanose ring. Biochim. Biophys. Acta, 1981. View Source
• [3] Transition-state discrimination by adenosine deaminase from Aspergillus oryzae. Biochemistry, 1993. 5′-deoxyadenosine and adenosine have nearly identical kcat/Km values. View Source
• [4] Adenosine deaminase prefers a distinct sugar ring conformation for binding and catalysis: kinetic and structural studies. Biochemistry, 2000. View Source

5′-Phosphorylation Blockade via Adenosine Kinase

Adenosine kinase (AK) requires a 5′-hydroxyl group on the ribofuranose (or xylofuranose) sugar for phosphorylation to the 5′-monophosphate [1]. Adenosine (Km for human AK ≈ 0.3–2 µM) and 9-(β-D-xylofuranosyl)adenine (xylo-A) are both substrates for AK and are efficiently converted to their respective 5′-triphosphates in CHO cells [2]. The 5′-deoxy modification in CAS 72-90-2 eliminates this metabolic activation pathway entirely, preventing 5′-nucleotide formation [1]. This metabolic blockade is not shared by adenosine or xylo-A and only partially by 5′-deoxyadenosine, which retains the β-D-ribo configuration and is cleaved to adenine by nucleoside phosphorylases rather than phosphorylated [3]. CAS 72-90-2 is thus predicted to be metabolically orthogonal: resistant to both 5′-phosphorylation (due to 5′-deoxy) and to phosphorolytic cleavage at rates comparable to 5′-deoxyadenosine, while the α-configuration may further reduce recognition by nucleoside transporters.

Evidence Dimension	Susceptibility to 5′-phosphorylation by adenosine kinase
Target Compound Data	CAS 72-90-2: 5′-phosphorylation not possible (no 5′-OH); α-configuration may additionally reduce AK binding. No published AK Km/IC₅₀ available.
Comparator Or Baseline	Adenosine: Km (human AK) ≈ 0.3–2 µM, efficiently phosphorylated to AMP. Xylo-A (9-(β-D-xylofuranosyl)adenine): readily phosphorylated to xylo-ATP in CHO cells. 5′-Deoxyadenosine: not phosphorylated (no 5′-OH) but cleaved to adenine by phosphorolytic enzymes in CHO, HeLa, and Novikoff hepatoma cells.
Quantified Difference	Qualitative: complete blockade of 5′-phosphorylation for 5′-deoxy compounds (CAS 72-90-2 and 5′-deoxyadenosine) vs. efficient phosphorylation for adenosine and xylo-A. Quantitative kinetic data for CAS 72-90-2: not yet published.
Conditions	Human adenosine kinase (recombinant); CHO cell metabolism studies; HeLa, Novikoff hepatoma, L929, L1210, P388 cell extracts

Why This Matters

For applications requiring a non-phosphorylatable adenine nucleoside probe—such as studying adenosine receptor signaling without confounding nucleotide second messenger effects, or designing metabolically stable nucleoside analogs—CAS 72-90-2 provides a definitive blockade of the 5′-phosphorylation route that adenosine and xylo-A cannot offer.

• [1] Long, M.C. et al. Structure-activity relationship for adenosine kinase from Mycobacterium tuberculosis II: Modifications to the ribofuranosyl moiety. Biochem. Pharmacol., 2008, 75, 1568–1581. View Source
• [2] Metabolism of 9-β-D-xylofuranosyladenine by the Chinese hamster ovary cell. Cancer Res. Xylo-A is phosphorylated to the triphosphate in CHO cells. View Source
• [3] 5′-Deoxyadenosine metabolism in various mammalian cell lines. Biochem. Pharmacol. 5′-dAdo cleaved to adenine in CHO, HeLa, Novikoff cells; not cleaved in L929, L1210, P388 cells. View Source

Anomeric Configuration Drives Antiviral Activity Profile

In the systematic evaluation of α- and β-D-xylofuranosyl nucleosides by Gosselin et al. (1986), the β-anomers—specifically 9-(β-D-xylofuranosyl)adenine (xylo-A), 9-(β-D-xylofuranosyl)guanine, and 1-(β-D-xylofuranosyl)cytosine—showed marked antiviral and cytostatic activity across a panel of RNA and DNA viruses [1]. The corresponding α-anomers were significantly less active or inactive in the same antiviral assays, establishing that the anomeric configuration is a critical determinant of biological recognition for xylofuranosyl nucleosides [1]. This anomer-dependent activity profile establishes a clear differentiation basis for CAS 72-90-2: its α-configuration predicts distinct enzyme/receptor interactions relative to xylo-A (β-D-xylofuranosyl, 5′-OH), and the combination of α-anomeric linkage with 5′-deoxy modification is unique among commercially available adenine nucleoside analogs. Independent studies on α-lyxofuranosyladenine have confirmed activity against herpes simplex virus types 1 and 2 both in vitro and in vivo, further supporting the principle that α-configured pentofuranosyl adenines can access biological targets inaccessible to their β-counterparts [2].

Evidence Dimension	Antiviral activity as a function of anomeric configuration in xylofuranosyl nucleosides
Target Compound Data	CAS 72-90-2 (α-D-xylofuranosyl, 5′-deoxy): no published antiviral data. Class-level inference from α-D-xylofuranosyl nucleosides: α-anomers generally less active than β-anomers in Gosselin 1986 panel.
Comparator Or Baseline	9-(β-D-xylofuranosyl)adenine (xylo-A): marked antiviral activity against multiple RNA and DNA viruses (Gosselin 1986). 9-(β-D-xylofuranosyl)guanine and 1-(β-D-xylofuranosyl)cytosine: also markedly active. α-D-xylofuranosyladenine: significantly less active or inactive in the same panel.
Quantified Difference	Qualitative: β-anomers active; α-anomers substantially less active in Gosselin 1986 antiviral panel. Reference compound: α-lyxofuranosyladenine active against HSV-1 and HSV-2 in vitro and in vivo.
Conditions	Antiviral testing against a variety of RNA and DNA viruses; cell growth inhibition and macromolecule synthesis assays (Gosselin 1986)

Why This Matters

CAS 72-90-2 serves as a stereochemically defined negative-control probe or scaffold for investigating how the α-anomeric configuration and 5′-deoxy modification jointly modulate biological activity, enabling structure-activity relationship (SAR) studies that are impossible with any single comparator analog.

• [1] Gosselin, G.; Bergogne, M.C.; de Rudder, J.; De Clercq, E.; Imbach, J.L. Systematic synthesis and biological evaluation of α- and β-D-xylofuranosyl nucleosides of the five naturally occurring bases in nucleic acids and related analogues. J. Med. Chem., 1986, 29, 203–213. View Source
• [2] De Rudder, J. et al. Antiviral activity of α-D-lyxofuranosyladenine against herpes simplex virus types 1 and 2 in vitro and in vivo. Cited in Gosselin 1986 and related patent literature. View Source

Chemical Building Block for Modified Oligonucleotide Synthesis

9-(5-Deoxy-α-D-xylofuranosyl)adenine (CAS 72-90-2) and its β-anomer serve as critical starting materials for synthesizing 5′-deoxy-xylonucleoside phosphoramidites used in solid-phase oligonucleotide synthesis [1]. The 5′-deoxy modification eliminates the need for 5′-O-dimethoxytrityl protection, simplifying the phosphoramidite synthesis workflow, while the xylofuranose sugar imparts distinct duplex stability and nuclease resistance properties to the resulting oligonucleotides [1]. In contrast, adenosine and 5′-deoxyadenosine (both β-D-ribofuranose) produce ribo- or 5′-deoxyribo-oligonucleotides with different hybridization thermodynamics, and xylo-A (β-D-xylofuranose, 5′-OH) requires additional 5′-protection/deprotection steps [2]. Furthermore, 9-(2′-deoxy-β-D-xylofuranosyl)adenine building blocks have been specifically developed for solid-phase synthesis of oligo(2′-deoxy-xylonucleotides), highlighting the distinct and non-interchangeable role of xylofuranosyl-configured adenine nucleosides in oligonucleotide chemistry [1].

Evidence Dimension	Synthetic utility for modified oligonucleotide preparation
Target Compound Data	CAS 72-90-2: α-configuration, 5′-deoxy simplifies phosphoramidite synthesis (no 5′-protection needed); xylofuranose sugar alters duplex properties
Comparator Or Baseline	Adenosine: requires 5′-O-DMTr protection, 2′-OH protection (TBDMS/TOM); produces RNA-type duplexes. 5′-Deoxyadenosine: no 5′-protection needed, β-D-ribo sugar. Xylo-A: requires 5′-O-DMTr protection, β-D-xylo sugar
Quantified Difference	Process advantage: elimination of 5′-protection/deprotection steps (saves 2 synthetic steps per coupling cycle). Structural advantage: xylofuranose vs. ribofuranose sugar produces distinct duplex stability and nuclease resistance profiles.
Conditions	Standard solid-phase oligonucleotide synthesis using phosphoramidite chemistry; DMTr = 4,4′-dimethoxytrityl; TBDMS = tert-butyldimethylsilyl

Why This Matters

For laboratories synthesizing modified oligonucleotides with altered pharmacokinetic or hybridization properties, CAS 72-90-2 offers a synthetically streamlined entry point (no 5′-protection required) combined with the unique conformational properties of the xylofuranose sugar, which cannot be replicated using ribofuranose-based adenosine analogs.

• [1] Seela, F. et al. 9-(2′-Deoxy-β-D-xylofuranosyl)adenine Building Blocks for Solid-Phase Synthesis and Properties of Oligo(2′-deoxy-xylonucleotides). Bioconjugate Chem., 1998. View Source
• [2] Gosselin, G.; Imbach, J.L. Nucléosides de synthèse XXII: Obtention, comme synthons oligoxylonucléotidiques, de D-xylofurannosyl adenines spécifiquement substitués. J. Heterocycl. Chem., 1982, 19, 597–602. View Source

High-Impact Application Scenarios

Anti-Tubercular Drug Discovery: LigA-Selective Inhibitor Libraries

Investigators developing inhibitors of Mycobacterium tuberculosis NAD⁺-dependent DNA ligase (LigA) require the 5-deoxy-α-D-xylofuranosyl head group as the essential recognition element for discriminating the bacterial enzyme from human ATP-dependent DNA ligase I. Using CAS 72-90-2 as the starting nucleoside, medicinal chemistry teams can elaborate focused libraries of bis-xylofuranosylated diamines with variable linkers (aminoalkyl or phenylene carbamoyl spacers) to optimize potency (target Ki < 10 µM) and selectivity (target > 10-fold over human LigI), building on the validated Compound 1 template (Ki = 72.8 µM; 9-fold selectivity) [1]. Neither adenosine, 5′-deoxyadenosine, nor xylo-A can serve as the starting material for this chemotype because each lacks the stereochemical signature (α-D-xylofuranosyl, 5′-deoxy) required for LigA active-site complementarity.

ADA Mechanistic Enzymology: Probing Sugar Conformation Contributions

The unique dual substrate-inhibitor behavior of the 5-deoxy-xylofuranosyladenine scaffold makes CAS 72-90-2 a valuable tool for dissecting the contributions of sugar puckering, anomeric configuration, and 5′-substitution to ADA catalysis. The β-anomer (6-amino-9-(5-deoxy-β-D-xylofuranosyl)purine) is both deaminated and inhibitory toward ADA [2], whereas adenosine is solely a substrate (Km ≈ 12 µM) and 5′-deoxyadenosine is solely a substrate with nearly identical catalytic efficiency [3]. Comparative kinetic studies using the α-anomer (CAS 72-90-2), β-anomer, adenosine, and 5′-deoxyadenosine can deconvolute the individual contributions of anomeric configuration, sugar stereochemistry, and 5′-hydroxyl presence to ADA binding affinity, catalytic rate, and inhibitory potency—parameters critical for designing ADA-resistant adenosine receptor ligands and chemotherapeutic nucleoside analogs.

Metabolically Stable Adenosine Receptor Probe Development

Pharmacologists studying adenosine receptor (A1, A2A, A2B, A3) signaling require ligands that bind the receptor without undergoing intracellular phosphorylation to nucleotides, which would confound readouts by activating P2 purinergic receptors or feeding into adenine nucleotide pools. CAS 72-90-2 is structurally incapable of 5′-phosphorylation due to the 5′-deoxy modification [4], in contrast to adenosine (efficiently phosphorylated; Km for AK ≈ 0.3–2 µM) and xylo-A (phosphorylated to xylo-ATP in CHO cells) [5]. This metabolic blockade, combined with the xylofuranose sugar's potential to bias the furanose ring conformation (affecting receptor subtype selectivity), positions CAS 72-90-2 as a scaffold for developing 'signaling-only' adenosine receptor probes that cleanly separate receptor activation from nucleotide metabolism.

Modified Oligonucleotide Therapeutics with Enhanced Nuclease Resistance

Oligonucleotide chemists developing therapeutic antisense or siRNA molecules with improved serum stability and target affinity can use CAS 72-90-2 to prepare 5′-deoxy-α-D-xylofuranosyl phosphoramidite building blocks for solid-phase synthesis. The 5′-deoxy modification streamlines synthesis by eliminating 5′-O-DMTr protection/deprotection cycles, while the xylofuranose sugar (3′-epimer of ribose) alters duplex geometry, thermal stability, and exonuclease susceptibility compared to standard ribose- or 2′-deoxyribose-based backbones [6]. The α-configuration further distinguishes the resulting oligonucleotides from those prepared with β-D-xylo building blocks (derived from xylo-A), enabling exploration of anomer-dependent effects on target mRNA binding and RNase H recruitment.

• [1] Srivastava, S.K. et al. Mycobacterium tuberculosis NAD⁺-dependent DNA ligase is selectively inhibited by glycosylamines compared with human DNA ligase I. Nucleic Acids Res., 2005, 33, 7090–7101. View Source
• [2] Shah, R.H.; Schaeffer, H.J.; Murray, D.H. Enzyme Inhibitors IV: Syntheses of 6-Substituted-9-(5-deoxy-β-D-xylofuranosyl)purines and Their Evaluation as Inhibitors of Adenosine Deaminase. J. Pharm. Sci., 1965, 54, 15–20. View Source
• [3] Transition-state discrimination by adenosine deaminase from Aspergillus oryzae. Biochemistry, 1993. View Source
• [4] Long, M.C. et al. Structure-activity relationship for adenosine kinase from Mycobacterium tuberculosis II. Biochem. Pharmacol., 2008, 75, 1568–1581. View Source
• [5] Metabolism of 9-β-D-xylofuranosyladenine by the Chinese hamster ovary cell. Cancer Res. View Source
• [6] Seela, F. et al. 9-(2′-Deoxy-β-D-xylofuranosyl)adenine Building Blocks for Solid-Phase Synthesis. Bioconjugate Chem., 1998. View Source