Prieurianin group

Certain compounds of the prieurianin group were reported to be separable by t.l.c. in Glasgow, but to isomerise on attempted separation in Durban (35, 136). This was eventually traced as an effect caused by the thickness of the absorbent layer. Thin plates, which run rapidly, separate the mixtures while thicker plates, which run more slowly, isomerise them. Not all commercially-available plates are suitable for these separations. Particular difficulty was experienced with the limonoids of Trichilia emetica. These proved inseparable by conventional methods 200) but were separated by a special technique of high pressure liquid chromatography 138). Because of this, some compounds which have been isolated may be artefacts a case in point is the limonoids of Aphanamixispolystacha (55) which were isolated only after repeated column chromatography. 

Some limonoids have also been found to be active against some types of cancer 108) to date activity is confined to certain compounds of the havanensin and prieurianin groups containing 14,15 epoxide rings. It is as yet too soon to say if the effect will be significant if it is, obtaining a supply of the rather rare and complex compounds involved will not be easy. 

Bicarbonate hydrolysis removed first the 11- and then the 12-substituent it was shown by mass spectroscopy that heudelottin E (18) had a formate at C-11, a 2-hydroxy-3-methylvalerate at C-12 and a 2-hydroxy-3-methylbutyrate at C-7. Heudelottin F (19) was acetylated in the side chain at C-12, while heudelottin C (20) was the 11-deformyl derivative of F. These compounds are of interest as they are the simplest ones containing the 11P-formyloxy-12oc-(2-hydroxy-3-methylvaleryloxy) system which has been found very commonly in the complex compounds of the prieurianin group. Acid treatment of heudelottin, like that of the simpler havanensin, very readily isomerises the oxide to a 15-ketone (22), but by alkaline hydrolysis of heudelottin it is possible to obtain the epoxy alcohol (17)... 

A similar compound, deacetyl-2-deoxyekebergin (42 a), is produced synthetically (76) as a by-product in the alkaline hydrolysis of the la,3a-diacetate corresponding to andirobin (Scheme 8). The mechanism of this change is obscure, since it takes place without inversion at C-14 possibly stabilisation of an intermediate by the methylene group at C-8 is involved. Whatever it may be, it seems reasonable that the mechanisms of the natural and the synthetic processes are similar and it may be significant that the 14p,15P-diols in the prieurianin series also have a methylene group at C-8. 

Prieurianin (90) (28) and dregeanin (91) (]86) were the first two members of this group to be described in 1965. Similar compounds have been found in the closely related genera Trichilia and Guarea, in Aphanamixis, and in Nymania. Most have the lip-formyl,12a-(3 -methyl-2 -hydroxy)valerate system which also occurs in simpler Trichilia compounds such as heudelottin E (18). 

Dregeanin and two y-lactones (103) obtained from it on hydrolysis, to be described further below, show an i.r. maximum near 1790 cmwhile D5 and the two rohituka substances show maxima at 1775 cm The reason for this is not known. It is possible that D5 may have the isomeric structure (104) with a 8-lactone and a tertiary hydroxy group. The arguments against this alternative structure are that all six of these compounds show a C-4 resonance at about 8 90, distinct from any other dregeanin derivative, which has been explained on the basis that C-4 is included in a y-lactone ring, (755). Also D5 and the rohituka compounds show a shift in the 2H-29 resonance from ca. 8 3.8 (s) to 8 4.1,4.3 (d, J = 12 Hz.) on acetylation, which is explained as due to the acetylation of a primary alcohol. Prieurianin has the C-4 resonance at 8 84.6, rohitukin, a 7,29 lactone, at 8 79.4 ppm. 

---