Potentially Prebiotic Passerini-Type Reactions of Phosphates

Abstract

Monoalkyl phosphates can take the place of carboxylic acids in the Passerini reaction, but excesses of aldehyde and isonitriles are required, and transfer of the phosphate to the newly formed hydroxyl group does not take place. By rendering the phosphate addition intramolecular, the efficiency of the reaction is substantially increased. Phosphate transfer can take place when cyanovinyl phosphate is used rather than a monoalkyl phosphate.

Key words

multicomponent reactions - Passerini reaction - phosphates - phosphorylation - prebiotic chemistry

References and Notes

• 1 Baldwin JE. Chan RY. Sutherland JD. Tetrahedron Lett.  1994,  35:  5519 
• 2 Mullen LB. Sutherland JD. Angew. Chem. Int. Ed.  2007,  46:  8063 
• 3 Pirrung MC. Das Sarma K. Tetrahedron  2005,  61:  11456 
• 4 Subsequent to our demonstration that alkyl phosphates can act as organocatalysts of the three-component Ugi reaction (ref. 2), List has identified phenyl phosphinic acid ‘as the best catalyst for this perfectly atom-economic reaction, thus introducing a new motif for organocatalysis’: Pan SC. List B. Angew. Chem. Int. Ed.  2008,  47:  3622 
• 6a Ferris JP. Science  1968,  161:  53 
• 6b Ferris JP. Goldstein G. Beaulieu DJ. J. Am. Chem. Soc.  1970,  92:  6598 
• 7 Krishnamurthy R. Guntha S. Eschenmoser A. Angew. Chem. Int. Ed.  2000,  39:  2281 

5

Glycolaldehyde phosphate disodium salt 8 (55 mg, 0.3 mmol) was dissolved in D₂O (3 mL) and the pD was adjusted to 6 using NaOD-DCl solution. Methyl isonitrile (18 µL, 0.3 mmol) was then added and the solution stirred at r.t. overnight.
Data for 11
^¹H NMR (500 MHz, D₂O): δ = 2.73 (s, 3 H, CH₃), 3.96 (ddd, J = 11.4, 7.3, 5.0 Hz, 1 H, CHHOP), 4.02 (ddd, J = 11.0, 6.3, 2.8 Hz, 1 H, CHHOP), 4.25 (app. t, J = 4.1, 3.8 Hz, 1 H, CHOH). ^¹³C NMR (75 MHz, D₂O): δ = 25.6 (CH₃), 66.3 (CH₂OP), 71.6 (CHOH), 174.3 (CONHCH₃). ^³¹P NMR (81 MHz, D₂O): δ = 1.32 (t, J = 6.4 Hz).

8

Glycolaldehyde 19 (15 mg, 0.25 mmol) and disodium cyanovinylphosphate (18, 145 mg, 0.75 mmol) were dissolved in D₂O (3 mL) and the pD was adjusted to 6 using NaOD-DCl solution. Methyl isonitrile (14 µL, 0.25 mmol) was then added and the solution stirred at r.t. overnight.
Data for 20
^¹H NMR (500 MHz, D₂O): δ = 2.74 (s, 3 H, CH₃), 3.81 (dd, J = 12.0, 3.2 Hz, 1 H, CHHOH), 3.85 (dd, J = 12.3, 2.8 Hz, 1 H, CHHOH), 4.44 (app. dt, J = 9.6, 3.2, 2.8 Hz, 1 H, CHOP). ^¹³C NMR (75 MHz, D₂O): δ = 25.7 (CH₃), 63.3 (CH₂OH), 75.2 (CHOP). ^³¹P NMR (81 MHz, D₂O): δ = 2.14 (d, J = 8.9 Hz). ESI-MS: m/z = 198 (87) [M^-]. HRMS: m/z calcd for C₄H₉O₆NP [M^-]: 198.0173; found: 198.0181.
Data for 21 ^¹H NMR (500 MHz, D₂O): δ = 2.72 (s, 3 H, CH₃), 3.72 (dd, J = 11.7, 4.9 Hz, 1 H, CHHOH), 3.76 (dd, J = 12.0, 3.5 Hz, 1 H, CHHOH), 4.16 (dd, J = 4.6, 3.6 Hz, 1 H, CHOH). ^¹³C NMR (75 MHz, D₂O): δ = 25.6 (CH₃), 63.3 (CH₂OH), 72.3 (CHOH). ESI-MS: m/z = 142 (100) [M + Na⁺].

9

It is likely that the cyanoacetaldehyde generated from 18 underwent aldol dimerisation (ref. 6). The resultant co-product would not have been clearly detectable by ^¹H NMR analysis because of (partial) hydrogen-deuterium exchange.