Quantification of gluconeogenesis (GNG) with carbon labeled tracers results in underestimation due to exchange and loss of tracer C in TCA cycle. We have reported the use of deuterium labeling of body water pool in order to label the glucogenic precursor phosphoenol pyruvate (JCI 95:172, 1995). The appearance of 2H on C-6 of glucose quantifies GNG from pyruvate, if correctred for incomplete equilibration of label at the pyruvate level, but does not include glycerol. Since 2H from water labels C-5 of glucose via pyruvate, and also labels C-2 of the triose phosphate pool, appearance of2 H on C-5 of glucose gives an estimate of total GNG, eliminating the uncertainty of incomplete equilibration and including glycerol.

This approach was applied to quantify GNG in normal pregnant women. Glucose Ra was measured using [U-13C]glucose tracer and GNG was quantified using [2H2]O tracer. [2H] on glucose C-5 was measured by first converting it to xylose and then cleaving C-5 to form formaldehyde which was converted to hexamethylenetetramine. Table

Table 1
Full size table

[2H] enrichment of glucose C-6 when compared with that in total body water (av 0.46%) suggests a greater contribution of GNG from pyruvate after an overnight fast in early than late pregnancy, 42% compared to 33%, assuming similar extents of equilibration between 2H in water and in pyruvate. Total GNG estimated from 2H on glucose C-5 was 56% of Ra in both the 1st and 3rd trimesters. The reported method allows meaurements of GNG without the uncertainties related to C-labeled tracers, since the precursor enrichment, 2H in body water is known, and dilution of label in the TCA cycle is not a factor.