“…13: TLC Rf 0.27 (ethyl acetate/toluene, 1:8); [a]16D +6.3°( c 1.08, CHClg); IR wCHC '3 3000, 2870, 1735, 1450, 1365, 1240, 1085 cm'1; XH NMR 1.60-2.08 (3 H, m, H-1,6,6'), 1.97, 2.01 ( 3X 2, each s, 2 x OCOCHg), 3.26 (1 H, dd, J = 10 and 3.5 Hz, H-3), 3.20-3.47 (1 H, m, H-5), 3.98 (2 H, d, J = 6 Hz, CH2OAc), 4.10 (1 H, t ,J= 3.5 Hz, H-4), 4.42-4.88 (6 H, m, 3 X OCH2C6H5), 5.35 (1 H, t, J = 9 Hz, H-2), 7.20-7.48 (+)-Codonopsinine [(+)-lb], the enantiomer of natural (-)-codonopsinine, and its stereoisomers, (2ñ,3S,4S,5S)-la, Codonopsinine and codonopsine, a new class of the 1,2,3,4,5-pentasubstituted pyrrolidine alkaloids isolated1,2 from Codonopsis clematidea, have been shown to have structures assigned as 1 and 2, respectively, by Russian workers. 3,4 In animal tests the latter has been found to possess hypotensive pharmacological activity with no effect on the central nervous system.5 Shortly after the structure elucidation, in 1972 the Russian group6 reported the relative stereochemistry for these alkaloids to be 2R*,3S*,4S*,5S* as shown in la and 2a, based on analyses of XH NMR coupling constants using the Karplus equation. However, vicinal coupling constants have been shown to be unreliable for assigning configurations of substituted pyrrolidines.7 Thus, it seemed to us that the proposed assignments rested on dubious spectral interpretation, and hence, additional verification was desirable.…”