Quantitative exposure assessments made using biologically relevant mar
kers will facilitate epidemiological studies of risk from environmenta
l carcinogens. Blood proteins are readily accessible macromolecules th
at have been shown to be targets for activated chemical carcinogens. S
erum albumin is quantitatively the most abundant target for aflatoxin
B1 and the measurement of aflatoxin-serum albumin adducts has been use
d to detect exposed individuals. The goal of these experiments was to
devise an analytical procedure that would increase the overall recover
y of aflatoxin adducts in serum albumin, and thereby improve the accur
acy of exposure monitoring. The method developed consisted of the foll
owing procedures. Proteins were precipitated from serum (less-than-or-
equal-to 100 mul) with 80% ammonium sulfate, with incubation at 4-degr
ees-C for 2 h. Following dialysis against phosphate-buffered saline (p
H 7.0 for 3 h at 4-degrees-C), the proteins were digested with proteas
e (Pronase) (1:4.1 w/w enzyme:protein) for 15 h at 37-degrees-C with s
haking. Enzyme and other undigested proteins were precipitated with ac
etone (1:2 v/v, 40 min, 4-degrees-C). After evaporation of the acetone
under vacuum, levels of aflatoxin B1-albumin adducts were determined
by radioimmunoassay carried out on 300 mul fractions. This procedure o
bviated the isolation of albumin prior to analysis and reduced interfe
rence in the radioimmunoassay. High recoveries of aflatoxin B1 adducts
were achieved together with a low limit of detection. The applicabili
ty of the procedure in epidemiological studies of human aflatoxin expo
sure was illustrated by results of analyis of aflatoxin-albumin adduct
s in serum samples from residents of Chongming Island, People's Republ
ic of China.