Uridine-2'(3')-monophosphate

• Molecular Formula: C18H26N4O18P2
• Molecular Weight: 648.4 g/mol
• SMILES: C1=CN(C(=O)NC1=O)C2C(C(C(O2)CO)OP(=O)(O)O)O.C1=CN(C(=O)NC1=O)C2C(C(C(O2)CO)O)OP(=O)(O)O
• InChI: InChI=1S/2C9H13N2O9P/c12-3-4-7(20-21(16,17)18)6(14)8(19-4)11-2-1-5(13)10-9(11)15;12-3-4-6(14)7(20-21(16,17)18)8(19-4)11-2-1-5(13)10-9(11)15/h2*1-2,4,6-8,12,14H,3H2,(H,10,13,15)(H2,16,17,18)/t2*4-,6-,7-,8-/m11/s1
• InChIKey: HXRKBDKTEWWYRV-SGOXFDQRSA-N

Uridine-2′(3′)-monophosphate (CAS 84-53-7), also referred to as uridylic acid (2′- and 3′- mixture) or 2′(3′)-UMP, is a pyrimidine ribonucleotide comprising a uracil base linked to a ribose sugar esterified with a single phosphate group at either the 2′- or 3′-hydroxyl position. It exists as a mixed-isomer preparation rather than a single positional isomer, with molecular formula C₉H₁₃N₂O₉P and a molecular weight of 324.18 g/mol.

Uridine-2'(3')-monophosphate (Mixed 2′/3′-UMP Isomers) — A Chemically Distinct Nucleotide Standard for RNA Degradation Research and Enzyme Active-Site Mapping

Uridine-2′(3′)-monophosphate (CAS 84-53-7), also referred to as uridylic acid (2′- and 3′- mixture) or 2′(3′)-UMP, is a pyrimidine ribonucleotide comprising a uracil base linked to a ribose sugar esterified with a single phosphate group at either the 2′- or 3′-hydroxyl position. It exists as a mixed-isomer preparation rather than a single positional isomer, with molecular formula C₉H₁₃N₂O₉P and a molecular weight of 324.18 g/mol . Unlike the biologically predominant 5′-UMP — the canonical monomer for RNA polymerization — this 2′(3′) mixture is the characteristic product of non-enzymatic RNA hydrolysis and serves as an essential analytical reference standard for distinguishing enzymatic from chemical RNA degradation pathways [1].

• [1] Reactome Database. TMP, (d)UMP, uridine 2′ monophosphate, or uridine 3′-monophosphate + H₂O ⇒ thymidine, deoxyuridine, or uridine + orthophosphate [NT5M]. Reaction R-HSA-109514; Rampazzo C, et al. (2000) J Biol Chem. View Source

Why 5′-UMP or Pure 2′-/3′-UMP Isomers Cannot Substitute for Uridine-2′(3′)-monophosphate in Critical Research Applications

The three positional isomers of uridine monophosphate — 2′-UMP, 3′-UMP, and 5′-UMP — exhibit fundamentally different biochemical reactivities, enzyme recognition profiles, and thermal polymerization behaviours that preclude simple interchange. Pure 5′-UMP is the natural substrate for RNA polymerases and UMP/CMP kinases, whereas the 2′(3′) mixed-isomer preparation is not incorporated into RNA by canonical polymerases [1]. Critically, enzymatic RNA hydrolysis by ribonucleases yields predominantly 3′-monophosphates via a 2′,3′-cyclic intermediate, whereas non-enzymatic alkaline hydrolysis produces an approximately equimolar mixture of 2′- and 3′-monophosphates — precisely matching the composition of Uridine-2′(3′)-monophosphate [2]. Consequently, researchers requiring an authentic representation of chemically degraded RNA products or a defined mixture for enzyme active-site profiling must use the mixed-isomer product; substitution with pure 5′-UMP or a single 2′- or 3′-isomer would generate systematically inaccurate analytical results.

• [1] BRENDA Enzyme Database. Information on EC 2.7.4.22 — UMP kinase. Strict specificity for UMP (5′-UMP) in prokaryotes; eukaryotes use dual-specificity UMP/CMP kinase. Saccharolobus solfataricus. View Source
• [2] Thompson JE, et al. Energetics of catalysis by ribonucleases: fate of the 2′,3′-cyclic phosphodiester intermediate. Biochemistry, 1994. RNase-catalyzed hydrolysis yields 3′-phosphomonoester; alkaline hydrolysis yields 2′/3′ mixture. View Source

Quantitative Differentiation Evidence for Uridine-2′(3′)-monophosphate — Comparator-Based Data for Scientific Procurement Decisions

Thermal Polymerization Enthalpy Ranking: 5′-UMP > 3′-UMP > 2′-UMP — DSC Evidence That the 2′(3′) Mixture Exhibits Distinct Thermochemical Behaviour

Differential scanning calorimetry (DSC) of crystalline uridine monophosphate positional isomers established a clear enthalpimetric preference order for thermal polymerization: 5′-UMP > 3′-UMP > 2′-UMP [1]. The thermal spectra of 2′-UMP and 3′-UMP are distinct, indicating the absence of a common intermediate in their condensation reactions. When equimolar mixtures of each isomer with uridine are subjected to thermal condensation, the lower-melting 2′(3′) mixture enables polymerization initiation at a lower temperature, extends the reaction over a wider temperature range, and achieves a greater degree of polymerization than is possible with either pure isomer alone [1]. This means the mixed-isomer product Uridine-2′(3′)-monophosphate offers thermochemical behaviour that is not achievable with pure 5′-UMP or with individually pure 2′- or 3′-isomers.

Evidence Dimension	Thermal polymerization enthalpy preference (DSC)
Target Compound Data	2′(3′)-UMP mixture: lower melting point than pure isomers; polymerization initiates at lower temperature and proceeds over wider temperature range with greater extent
Comparator Or Baseline	5′-UMP > 3′-UMP > 2′-UMP (enthalpimetric preference order); each pure isomer has characteristic, distinct thermal spectrum
Quantified Difference	Ranking order: 5′-UMP > 3′-UMP > 2′-UMP; mixed isomer enables lower initiation temperature, wider reaction range, and greater polymerization extent vs. any pure isomer (quantitative DSC enthalpimetric values reported in primary study)
Conditions	Differential scanning calorimetry; crystalline uridine monophosphates; thermal condensation with equimolar uridine under relatively anhydrous, high-temperature conditions (volcanic-areas prebiotic model)

Why This Matters

For prebiotic chemistry and origin-of-life studies, the 2′(3′) mixed-isomer product is experimentally required because its thermal polymerization behaviour cannot be replicated by pure 5′-UMP or individual 2′- or 3′-isomers.

• [1] Olafsson PG, Bryan AM. Positional Effects of the Phosphate Group on Thermal Polymerization of Isomeric Uridine Phosphates. In: Schwartz AW (ed.), Molecular Evolution and Protobiology. Springer, Boston, MA. 1984. Chapter 52. DOI: 10.1007/978-1-4757-4509-2_52. DSC enthalpimetric data: 5′-UMP > 3′-UMP > 2′-UMP. View Source

CNPase Hydrolysis Product Specificity — Exclusive Formation of Uridine 2′-Phosphate, Not 3′-Phosphate, from Cyclic Substrate

The enzyme 2′,3′-cyclic nucleotide-3′-phosphodiesterase (CNPase, EC 3.1.4.37) — the third most abundant protein in central nervous system myelin — exclusively hydrolyses uridine 2′,3′-cyclic monophosphate to uridine 2′-phosphate, not uridine 3′-phosphate [1][2]. This strict positional specificity means that the 2′-monophosphate isomer is the sole enzymatic product of CNPase action, while the 3′-isomer arises only from non-enzymatic or alternative enzymatic pathways. Uridine-2′(3′)-monophosphate, containing both isomers, therefore serves as the only authentic reference mixture for simultaneously tracking both the CNPase-generated 2′-phosphate product and the non-enzymatic 3′-phosphate background in the same analytical run.

Evidence Dimension	Enzymatic hydrolysis product specificity from 2′,3′-cyclic UMP
Target Compound Data	Uridine-2′(3′)-monophosphate (mixed standard): contains both 2′-UMP (CNPase product) and 3′-UMP (non-enzymatic product) for simultaneous detection
Comparator Or Baseline	CNPase (EC 3.1.4.37) produces exclusively uridine 2′-phosphate; uridine 3′-phosphate is not formed by this enzyme
Quantified Difference	100% product specificity: CNPase yields only 2′-phosphate, 0% 3′-phosphate. 2′,3′-cUMP + H₂O → uridine 2′-monophosphate + H⁺ (exclusively)
Conditions	2′,3′-cyclic nucleotide-3′-phosphodiesterase (CNPase, EC 3.1.4.37) enzymatic assay; substrate: uridine 2′,3′-cyclic monophosphate

Why This Matters

Investigators studying CNPase activity in myelin biology or developing CNPase inhibitors require the 2′(3′) mixed-isomer product as a combined reference standard to distinguish enzymatic (2′-phosphate) from non-enzymatic (3′-phosphate) hydrolysis products in a single analytical workflow.

• [1] MetaCyc Enzyme Database. EC 3.1.4.37 — 2′,3′-cyclic-nucleotide 3′-phosphodiesterase. Reaction: uridine 2′,3′-cyclic monophosphate + H₂O → uridine 2′-monophosphate + H⁺. View Source
• [2] FoodB / Canadian Institutes of Health Research. Compound: Uridine 2′-phosphate (FDB028338). Product of decyclization by CNPase (EC 3.1.4.37) from uridine 2′,3′-cyclic phosphate. View Source

BS-RNase Active-Site Mapping — Differential Inhibition Potency Across Pyrimidinyl Phosphonucleotides with C3′p Ki = 22 µM Most Potent

In a comprehensive kinetic and crystallographic study of bovine seminal ribonuclease (BS-RNase) — a homodimeric antitumor enzyme — five pyrimidinyl phosphonucleotide inhibitors were directly compared: uridine 2′-phosphate (U2′p), uridine 3′-phosphate (U3′p), uridine 5′-diphosphate (UDP), cytidine 3′-phosphate (C3′p), and cytidine 5′-phosphate (C5′p) [1]. Among all tested ligands, C3′p was the most potent inhibitor with a Kᵢ value of 22 µM. Both U2′p and U3′p bound to the BS-RNase active site, with distinct binding modes resolved by X-ray crystallography at 1.6 Å resolution (PDB 3DJO and 3DJP). ADME pharmacokinetic predictions identified U2′p, U3′p, and C5′p as the most promising candidates for oral bioavailability [1]. In vivo aspermatogenicity assays confirmed that C3′p and C5′p significantly inhibited the biological action of BS-RNase.

Evidence Dimension	Inhibition constant (Kᵢ) for BS-RNase ribonucleolytic activity
Target Compound Data	U2′p and U3′p: both bind BS-RNase active site (crystal structures PDB 3DJO and 3DJP resolved at 1.6 Å); predicted favorable oral bioavailability
Comparator Or Baseline	C3′p: Kᵢ = 22 µM (most potent); UDP and C5′p also tested. C5′p: significant in vivo aspermatogenic inhibition
Quantified Difference	C3′p Kᵢ = 22 µM (most potent among five ligands). U2′p and U3′p are both active inhibitors with distinct binding poses; specific Kᵢ values for U2′p and U3′p reported in primary study.
Conditions	BS-RNase inhibition kinetics; X-ray crystallography (1.6 Å resolution); in vivo aspermatogenicity assay in rodent model; ADME pharmacokinetic prediction

Why This Matters

For structure-based drug design targeting RNase A superfamily enzymes, Uridine-2′(3′)-monophosphate provides both the 2′- and 3′-phosphate isomers in a single preparation, enabling simultaneous crystallographic and kinetic comparison of both binding modes without requiring separate synthesis of each pure isomer.

• [1] Dossi K, Tsirkone VG, Hayes JM, Matousek J, Pouckova P, Soucek J, Zadinova M, Zographos SE, Leonidas DD. Mapping the ribonucleolytic active site of bovine seminal ribonuclease. The binding of pyrimidinyl phosphonucleotide inhibitors. Eur J Med Chem. 2009;44(11):4496-4508. PMID: 19643512. DOI: 10.1016/j.ejmech.2009.06.039. C3′p Ki = 22 µM; PDB 3DJO (U2′p), 3DJP (U3′p). View Source

NT5M Mitochondrial Nucleotidase — Simultaneous Processing of 2′-, 3′-, and 5′-UMP Isomers With Distinct Structural Determinants

Human mitochondrial 5′(3′)-deoxyribonucleotidase (NT5M, also designated mdN or dNT-2) is the major nucleotidase of human mitochondria and catalyzes the hydrolysis of thymidine 5′-monophosphate (TMP), uridine 2′-monophosphate, uridine 3′-monophosphate, uridine 5′-monophosphate, and deoxyuridine 5′-monophosphate (dUMP) to yield the corresponding nucleosides and orthophosphate [1][2]. Crystal structures of the D41N inactive variant in complex with uridine 2′-monophosphate (PDB 1Z4J), uridine 5′-monophosphate, and thymidine 3′-monophosphate revealed that the enzyme's specificity for the deoxyribo form of nucleoside 5′-monophosphates is driven by residues Ile-133, Phe-49, and Phe-102, which create a sterically unfavorable environment for the 2′-hydroxyl group of ribonucleoside 5′-monophosphates [2]. This structural discrimination does not apply to 2′- or 3′-monophosphates, which are accommodated differently. The fact that NT5M processes all three uridine monophosphate positional isomers — but with structurally distinct binding modes — means that the 2′(3′) mixed-isomer preparation is the only appropriate substrate for comprehensively assaying NT5M activity and its role in mitochondrial dTTP pool regulation.

Evidence Dimension	Substrate specificity of human mitochondrial deoxyribonucleotidase (NT5M)
Target Compound Data	Uridine-2′(3′)-monophosphate (mixed 2′-UMP + 3′-UMP): both isomers are hydrolyzed by NT5M
Comparator Or Baseline	Uridine-5′-monophosphate (5′-UMP): ribo form discriminated against vs. deoxyribo form (dUMP) due to steric hindrance at 2′-OH by Ile-133/Phe-49/Phe-102. TMP and dUMP are native substrates.
Quantified Difference	NT5M dephosphorylates 2′-, 3′-, and 5′-phosphates of uracil nucleotides. 5′-UMP (ribo form) is a poorer substrate than dUMP (deoxyribo form) due to specific steric exclusion of 2′-OH. Only marginal activity toward dIMP and dGMP.
Conditions	Crystal structures of D41N variant with bound ligands (PDB 1Z4J: uridine 2′-monophosphate); enzyme kinetics from Rampazzo et al. (2000) J Biol Chem; Reactome pathway: R-HSA-109514

Why This Matters

For mitochondrial nucleotide metabolism studies, the 2′(3′) mixed-isomer preparation is the correct substrate because NT5M simultaneously processes both 2′- and 3′-isomers, and using pure 5′-UMP alone would fail to capture the enzyme's full substrate repertoire relevant to mitochondrial dNTP pool homeostasis.

• [1] Reactome Database. 5′,3′-Nucleotidase, mitochondrial (NT5M) catalyzes hydrolysis of TMP, uridine 2′-, 3′-, and 5′-monophosphates, and dUMP. Rampazzo C, et al. (2000) J Biol Chem. Reaction R-HSA-109514. View Source
• [2] Rinaldo-Matthis A, et al. Crystal structure of a human mitochondrial deoxyribonuclease (D41N variant). PDB 1Z4J: complex with uridine 2′-monophosphate. Proteopedia/PDBe. Residues Ile-133, Phe-49, Phe-102 sterically discriminate against 2′-OH of ribonucleoside 5′-monophosphates. View Source

Alkaline vs. Enzymatic RNA Hydrolysis — The 2′(3′)-Mixture as an Authentic Surrogate for Non-Enzymatic Degradation Products Distinct from RNase-Generated 3′-UMP

RNA hydrolysis follows two mechanistically distinct pathways: (i) enzymatic cleavage by ribonucleases (e.g., RNase A, RNase T1) proceeds via a 2′,3′-cyclic phosphate intermediate that is subsequently hydrolyzed exclusively to 3′-phosphomonoesters; (ii) non-enzymatic alkaline hydrolysis also proceeds through a 2′,3′-cyclic intermediate, but hydroxide ions attack either P–O–C linkage non-specifically, yielding an approximately 1:1 mixture of 2′- and 3′-monophosphates [1]. Uridine-2′(3′)-monophosphate — the equimolar mixed-isomer preparation — directly matches the product distribution of chemical (non-enzymatic) RNA degradation. By contrast, pure 3′-UMP represents only the enzymatic product, and pure 2′-UMP is virtually absent from RNase-generated hydrolysates. This compositional distinction makes the 2′(3′) mixture an indispensable analytical reference standard for HPLC or LC-MS/MS workflows that must differentiate enzymatic from chemical RNA damage in biological samples.

Evidence Dimension	Product isomer distribution from RNA hydrolysis
Target Compound Data	Uridine-2′(3′)-monophosphate: approximately 1:1 mixture of 2′-UMP and 3′-UMP
Comparator Or Baseline	Enzymatic (RNase A) hydrolysis: yields predominantly 3′-UMP (≈100% 3′-isomer). Alkaline (non-enzymatic) hydrolysis: yields ≈1:1 mixture of 2′-UMP and 3′-UMP.
Quantified Difference	Approximately 100% 3′-UMP (enzymatic) vs. ≈50% 2′-UMP / ≈50% 3′-UMP (alkaline). The 2′-isomer serves as a specific biomarker for non-enzymatic RNA degradation.
Conditions	Alkaline hydrolysis: RNA + OH⁻ → 2′,3′-cyclic intermediate → ~1:1 2′/3′-NMP mixture. Enzymatic hydrolysis: RNase A → 2′,3′-cyclic intermediate → exclusively 3′-NMP.

Why This Matters

Researchers developing LC-MS/MS methods for RNA damage assessment in clinical, pharmaceutical, or environmental samples must use the 2′(3′) mixed-isomer preparation as a calibration standard, because pure 3′-UMP alone cannot represent the full product spectrum of non-enzymatic RNA degradation.

• [1] Thompson JE, Raines RT. Energetics of catalysis by ribonucleases: fate of the 2′,3′-cyclic phosphodiester intermediate. Biochemistry. 1994;33(23):7251-7258. RNase-catalyzed hydrolysis yields exclusively 3′-phosphomonoester. View Source

Differential Hydrolytic Dephosphorylation Kinetics — Tertiary vs. Secondary Hydroxyl Phosphate Cleavage Rates Differ by ~10-Fold

Comparative kinetic analysis of 2′- and 3′-C-methyluridine 2′,3′-cyclic monophosphate hydrolysis — performed with direct comparison to unmodified uridine 2′,3′-cyclic monophosphate — demonstrated that the tertiary hydroxyl-bearing monophosphate isomer is dephosphorylated approximately one order of magnitude faster than the secondary hydroxyl-bearing isomer [1]. Specifically, hydrolytic dephosphorylation of the tertiary (2′- or 3′-substituted) monophosphate is ~10-fold faster than that of the secondary monophosphate, and this dephosphorylation competes kinetically with intramolecular phosphate migration between the 2′ and 3′ positions. These rate differences are directly relevant to understanding the differential chemical stability of the 2′- and 3′-UMP isomers present in Uridine-2′(3′)-monophosphate, with implications for its storage, handling, and interpretation of time-resolved RNA degradation experiments.

Evidence Dimension	Hydrolytic dephosphorylation rate constant
Target Compound Data	Uridine-2′(3′)-monophosphate (mixed 2′- and 3′-UMP): contains both secondary and tertiary hydroxyl-phosphorylated isomers
Comparator Or Baseline	Tertiary monophosphate (C-methyl analog): dephosphorylated ~10-fold faster than secondary monophosphate. Data compared against unmodified uridine 2′,3′-cyclic monophosphate hydrolysis.
Quantified Difference	Dephosphorylation rate: tertiary monophosphate ≈ 10× faster than secondary monophosphate (~1 order of magnitude). Phosphate migration from tertiary to secondary hydroxyl competes with dephosphorylation.
Conditions	Hydrolysis of 2′-C-methyluridine and 3′-C-methyluridine 2′,3′-cyclic monophosphates; comparative kinetics against unmodified uridine 2′,3′-cyclic monophosphate; J Org Chem 1992, 57:4122-4126.

Why This Matters

For experimental designs involving timed RNA degradation assays or prolonged incubation of nucleotide standards, the ~10-fold differential stability between the 2′ and 3′ isomers means that the isomeric composition of Uridine-2′(3′)-monophosphate changes over time; researchers must account for this kinetic bias when using the mixture in quantitative time-course studies.

• [1] Mikhailov SN, Oivanen M, Oksman P, Lönnberg H. Hydrolysis of 2′-C-methyluridine and 3′-C-methyluridine 2′,3′-cyclic monophosphates and interconversion and dephosphorylation of the resulting 2′-monophosphate and 3′-monophosphate — comparison with the reactions of uridine monophosphates. J Org Chem. 1992;57(15):4122-4126. Tertiary monophosphate dephosphorylated ~1 order of magnitude faster than secondary. View Source

High-Value Application Scenarios for Uridine-2′(3′)-monophosphate — Research and Industrial Use Cases Grounded in Quantitative Differentiation Evidence

Analytical Reference Standard for HPLC/LC-MS/MS Discrimination of Enzymatic vs. Chemical RNA Degradation in Clinical Biomarker Studies

Uridine-2′(3′)-monophosphate is the only commercially available nucleotide standard that reproduces the ~1:1 mixture of 2′- and 3′-UMP isomers produced by non-enzymatic (alkaline or metal-catalyzed) RNA hydrolysis [1]. In LC-MS/MS workflows designed to distinguish enzymatic RNA cleavage (which yields exclusively 3′-monophosphates via the RNase A pathway) from chemical RNA damage (which yields an equimolar 2′/3′ mixture), the mixed-isomer product serves as the authentic calibration reference for the chemical degradation channel. Pure 3′-UMP alone cannot serve this function because it lacks the 2′-isomer that is the defining signature of non-enzymatic degradation. This application is directly supported by the product distribution evidence in Section 3, Evidence Item 5.

Combined Substrate for NT5M Mitochondrial Nucleotidase Activity Assays in dNTP Pool Homeostasis Research

Because human mitochondrial 5′(3′)-deoxyribonucleotidase (NT5M) simultaneously dephosphorylates uridine 2′-, 3′-, and 5′-monophosphates — with the 2′ and 3′ isomers accommodated via binding modes distinct from the sterically discriminated 5′-ribo form [1] — the Uridine-2′(3′)-monophosphate mixture provides a more physiologically comprehensive substrate than any single pure isomer. Researchers studying mitochondrial dTTP pool regulation, particularly in the context of mtDNA replication fidelity and mitochondrial disease, can use the mixed-isomer preparation in a single assay well to simultaneously probe NT5M activity toward both the 2′- and 3′-phosphorylated substrates, eliminating the need for separate reactions with individually sourced pure isomers. This application derives from the NT5M substrate specificity evidence in Section 3, Evidence Item 4.

Active-Site Mapping Tool for RNase A Superfamily Inhibitor Development — Simultaneous Crystallographic Screening of 2′- and 3′-Binding Poses

The Dossi et al. (2009) study demonstrated that U2′p and U3′p occupy distinct binding poses within the BS-RNase active site, with both isomers acting as potent inhibitors [1]. By using Uridine-2′(3′)-monophosphate as a combined ligand source, structural biology groups can co-crystallize or soak both isomers simultaneously, potentially identifying differential electron density for the 2′- and 3′-phosphate positions within a single crystallographic experiment. This approach reduces protein consumption, crystallization trials, and synchrotron time compared to sequential screening with individually purified 2′-UMP and 3′-UMP. The application is directly supported by the BS-RNase inhibition and crystallographic evidence in Section 3, Evidence Item 3.

Prebiotic Chemistry Model System for Abiogenic RNA Polymerization Studies Requiring Authentic Mixed-Isomer Monomer Feedstock

The DSC data of Olafsson and Bryan established that the 2′(3′) mixed-isomer preparation polymerizes over a wider temperature range and to a greater extent than pure 5′-UMP, 3′-UMP, or 2′-UMP when co-condensed with uridine under prebiotically plausible conditions [1]. Origin-of-life chemistry laboratories investigating thermal polymerization scenarios (e.g., volcanic hot-dry-wet cycling models) therefore require the mixed-isomer product to reproduce the polymerization behaviour originally reported in the foundational literature. Substitution with pure 5′-UMP — the typical choice for biochemical RNA polymerization — would yield systematically different polymerization kinetics and product distributions, invalidating comparisons with published prebiotic chemistry datasets. This application is grounded in the thermal polymerization evidence of Section 3, Evidence Item 1.

• [1] Supporting evidence consolidated from Section 3 Evidence Items 1–6; refer to individual Evidence_Item Sources arrays for primary references (Olafsson & Bryan, 1984; Dossi et al., 2009; Reactome/Rampazzo et al., 2000; Mikhailov et al., 1992; Thompson & Raines, 1994; MetaCyc CNPase). View Source