Identification of the compound eliminated by the urine. - The fraction soluble in acetone from the urine of the experimental animals showed a blue-gray spot at 365 nm in TLC with an Rf of 0.28 that was not present in the urine of the control rats nor in the B. megalandra leaves aqueous extract. The blue-gray compound was partially purified by using column chromatography and differential solubilization and was partially identified by ¹H and ¹³C NMR.

In the ¹³C spectrum (Table 1) six signals between 122 and 110 ppm were observed and they were assigned to benzene carbons being two of them probably quaternary. Furthermore signals were observed at: 175.94 ppm that may correspond to a carboxylic acid, ester or amide; 162.29 ppm that correspond to the carbonyl carbon of the urea; a signal at 62.12 ppm and three signals of aliphatic carbons at 24.56, 19.67 and 13.10 ppm.

In the ¹H spectrum (Table 1) many signal were observed, however certain of them are of interest: four signals at low field, coupled between them at 7.7 and 6.9 ppm that correspond to aromatic protons and define the substitution pattern AA'BB'. Additionally, a broad and intense singlet was observed at 5.7 ppm that corresponds to the urea protons. At high field a triplet was observed at 1.23 ppm (J= 7.2 Hz) that is coupled to a quartet in 4.09 (J= 7.2 Hz) that defines as an ethoxy (EtO-), one of the substituent groups.

The coupling pattern in the aromatic region of the ¹H NMR coincides with a benzene ortho-substituted and it is explained thus: two doublets of doublets centered at 7.62 and 7.28 ppm with a coupling ortho-meta (J= 8; 8 and 2 Hz). Additionally, a singlet at 7.9 ppm was observed, possibly these signals can be attributed to an aromatic system poly-substituted. A possible structure that fits the majority of the NMR signals could be:

Changes in glycaemia and glycosuria during the administration of the B. megalandra leaf extract. – As it can be observed in Table 2 the administration of the plant extract produces a decrease of approximately 9 % in the glycaemia which was statistically significant at p< 0.01. On the other hand, the 24 h. glycosuria increases by approximately 3.6 times this being statistically significant at p< 0.005

The experimental rats received B. megalandra aqueous leaves extract in place of drinking water for 7 days. During a 24 h fast period the urine was collected, after which a sample of blood was taken, the glucose was measured in the plasma and urine by the GOD/POD ⁽¹⁰⁾. The values represent the mean ± standard deviation. The differences between controls and experimental was significant at p< 0.01 for the glycaemia and p< 0.005 for the glycosuria.