Hydroxypyruvate reductase (KPR) has been partially purified from human live
r and can be separated into at least two forms by chromatofocusing; these f
orms therefore differ in their pi values. Both forms, one with a pi of >7.2
(peak A) and the other with a pi between pH 6.5 and 5.5 (peak B), use NADP
H as a cofactor. However, only peak B was able to reduce hydroxypyruvate an
d glyoxylate, with a K-m, of 2.3 mM for the latter substrate. Peak A coelut
ed with lactate dehydrogenase and could represent lactate dehydrogenase (wh
ich is known to reduce hydroxypyruvate) alone or a mixture of proteins with
HPR activity. The K-m, for hydroxypyruvate of the enzyme(s) in peak A (8 m
M) was 80 times greater than that of peak B (0.1 mM), suggesting that the H
PR enzyme contained in peak B may be more important physiologically, where
the hydroxypyruvate concentrations are in the micromolar range. The data pr
esented provide a biochemical explanation for the previously observed diffe
rences in the tissue distribution of HPR and glyoxylate reductase activitie
s in human subjects and support the claim that diagnoses of primary hyperox
aluria type 2 should be made by measurement of glyoxylate reductase activit
y in the liver.